Phileducation John Dewey

INTERNATIONAL LIBRARY OF THE
PHILOSOPHY OF EDUCATION
JOHN DEWEY RECONSIDERED
JOHN DEWEY RECONSIDERED
Edited by
R.S.PETERS
Volume 19
LONDON AND NEW YORK
First published in 1977

This edition first published in 2010
by Routledge
2 Park Square, Milton Park, Abington, Oxon, OX14 4RN
Simultaneously published in the USA Canada
by Routledge
270 Madison Avenue, New York, NY 10016
Routledge is an imprint of the Taylor & Francis Group, an informa business
This edition published in the Taylor & Francis e-Library, 2010.
To purchase your own copy of this or any of
Taylor & Francis or Routledges collection of thousands of eBooks
please go to www.eBookstore.tandf.co.uk.
1977 Routledge & Kegan Paul Ltd
All rights reserved. No part of this book may be
reprinted or reproduced or utilised in any form or by
any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying
and recording, or in any information storage or
retrieval system, without permission in writing from
the publishers.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the
British Library
ISBN0-203-86104-3Master e-book ISBN
ISBN 10:0-415-55946-4 (Set)

eISBN 10:0-2038-6097-7 (Set)
ISBN 10:0-415-56252-X (Volume 19)
eISBN 10:0-203-86104-3 (Volume 19)
ISBN 13:978-0-415-55946-1 (Set)
eISBN 13:978-0-2038-6097-7 (Set)
ISBN 13:978-0-415-56252-2 (Volume 19)
eISBN 13:978-0-203-86104-2 (Volume 19)
Publishers Note
The publisher has gone to great lengths to ensure the
quality of this reprint but points out that some imperfections
in the original copies may be apparent.
Disclaimer
The publisher has made every effort to trace copyright
holders and would welcome correspondence
from those they have been unable to trace.
John Dewey reconsidered

Edited by
R.S.Peters
Routledge & Kegan Paul

London, Henley and Boston
First published in 1977

by Routledge & Kegan Paul Ltd
39 Store Street,
London WC1E 7DD,
Broadway House,
Newtown Road,
Henley-on-Thames,
Oxon RG9 1EN and
9 Park Street,
Boston, Mass. 02108, USA
This edition published in the Taylor & Francis e-Library, 2010.
To purchase your own copy of this or any of
Taylor & Francis or Routledges collection of thousands of eBooks
please go to www.eBookstore.tandf.co.uk.
Routledge & Kegan Paul Ltd 1977
No part of this book may be reproduced in
any form without permission from the
publisher, except for the quotation of brief
passages in criticism
British Library Cataloguing in Publication Data
John Dewey reconsidered.(International
library of the philosophy of education).
1. Dewey, JohnAddresses, essays, lectures
I. Peters, Richard Stanley II. Series
191 B945.D44 7730006
ISBN0-203-86104-3Master e-book ISBN
ISBN 0 7100 8623 7 (Print Edition)
Contents
General editors note
1 Inquiry, thought and action: John Deweys theory of knowledge

Anthony Quinton
2 Language and experience

Jerome Bruner, Eileen Caudill and Anat Ninio
3 Deweys theory of interest

Alan R.White
4 The self in action

Martin Hollis
5 Democracy and education

Antony Flew
6 John Deweys philosophy of education

R.S.Peters
Index
vii
1
12
23
36
48
65
79
General editors note
There is a growing interest in philosophy of education amongst students of philosophy as

well as amongst those who are more specifically and practically concerned with educational problems. Philosophers, of course, from the time of Plato onwards, have taken an
interest in education and have dealt with education in the context of wider concerns about
knowledge and the good life. But it is only quite recently in this country that philosophy of
education has come to be conceived of as a specific branch of philosophy like the philosophy of science or political philosophy.
To call philosophy of education a specific branch of philosophy is not, however, to
suggest that it is a distinct branch in the sense that it could exist apart from established
branches of philosophy such as epistemology, ethics, and philosophy of mind. It would be
more appropriate to conceive of it as drawing on established branches of philosophy and
bringing them together in ways which are relevant to educational issues. In this respect
the analogy with political philosophy would be a good one. Thus use can often be made of
work that already exists in philosophy. In tackling, for instance, issues such as the rights of
parents and children, punishment in schools, and the authority of the teacher, it is possible
to draw on and develop work already done by philosophers on rights, punishment, and
authority. In other cases, however, no systematic work exists in the relevant branches
of philosophye.g. on concepts such as education, teaching, learning, indoctrination. So philosophers of education have had to break new groundin these cases in the
philosophy of mind. Work on educational issues can also bring to life and throw new light
on long-standing problems in philosophy. Concentration, for instance, on the particular
predicament of children can throw new light on problems of punishment and responsibility.
G.E.Moores old worries about what sorts of things are good in themselves can be brought
to life by urgent questions about the justification of the curriculum in schools.
There is a danger in philosophy of education, as in any other applied field, of polarization to one of two extremes. The work could be practically relevant but philosophically
feeble, or it could be philosophically sophisticated but remote from practical problems.
The aim of the new International Library of the Philosophy of Education is to build up a
body of fundamental work in this area which is both practically relevant and philosophically competent. For unless it achieves both types of objective it will fail to satisfy those
for whom it is intended and fall short of the conception of philosophy of education which
the International Library is meant to embody.
John Dewey was a philosopher who pre-eminently tried to relate his philosophy to practical concerns. He was also best known for his philosophy of education. It is therefore
appropriate that an attempt to reappraise both his general philosophy of man and society
as well as his philosophy of education, which was intimately connected with both, should
appear in the International Library of the Philosophy of Education.
viii General editors note

This collection of papers was made possible by the generosity of the John Dewey Foundation who gave a grant to the University of London Institute of Education to put on a
course of public lectures on John Deweys philosophy, in the hope that the reading and
study of the works of John Dewey would thereby be encouraged. After the lectures there
were countless requests that they should be made available in published form to be read
and examined at more leisure. This collection is a response to such requests.
R.S.P.
1
Inquiry, thought and action:
John Deweys theory of knowledge
Anthony Quinton
1 Introduction
Pragmatism began as a theory of meaning. It is often dated from the publication in 1878 of
Peirces article How to make our ideas clear, in which the meaning of an idea is identified with its practical bearings, that is to say the difference its being true would make in
terms of experiencable consequences in the future. It is perhaps most familiar as a theory
of truth, especially in the form given to it by William James for whom the true is what it is
good, expedient or satisfactory to believe.
But pragmatism is also a theory of knowledge and was so from the beginning. In two
articles that appeared in W.T. Harriss Journal of Speculative Philosophy as early as 1868,
its second year (Questions concerning certain faculties claimed for man and Some consequences of four incapacities), Peirce argues elaborately and ingeniously against what
he calls Cartesianism, the idea that knowledge should be constructed from intuitively selfevident beliefs in minds that have been cleared of all habitual assumptions by a process of
universal doubt.
As Peirce himself says, most modern philosophers have been, in effect, Cartesians,
and this is true of British philosophers in particular. Lockes view that all knowledge must
be derived from the intuitive deliverances of sensation and reflection, if empirical, and
from intuitive awareness of conceptual connections, if it is not, has been the common
conviction of the central tradition in British epistemology that stems from him, through
Berkeley and Hume and then John Stuart Mill to Russell and such Russellian theorists of
knowledge of more recent times as Price and Ayer. What is more, the varying accounts the
members of this tradition have given, on their shared Cartesian basis, of our knowledge of
the material world, the minds of ourselves and others, of the past and of the laws of nature,
have remained, until the last few decades at least, the centre of constructive philosophical
interest and are still today the initial core of philosophical teaching.
Yet until the publication of W.B.Gallies Peirce and Pragmatism in 1952 there was no
British discussion of Peirces anti-Cartesianism and even now such currency as its conclusions have is perhaps more attributable to the influence of the rather similar account
of the basis-problem in Poppers Logic of Scientific Discovery than to Peirces presentation of it.
Jamess epistemological ideas are, indeed, familiar in the developed form of his radical empiricism through the fact that Russell was converted to them. In his essay on The
2 John Dewey reconsidered

nature of acquaintance in 1914, Russell criticized Jamess neutral monism from the point
of view of his conviction that in perception a mental subject is related to a physical object
that is totally different in nature from it. By 1918 with The philosophy of logical atomism
and, even more, by 1922 with the Analysis of Mind, Russells conversion to Jamess way of
thinking was complete. But this was hardly a conversion to pragmatism, even to pragmatist
epistemology, from which Jamess doctrine, with its close affinities to the ideas of Mach
and, on a natural reading, Hume, is really a deviation.
The chief continuator of Peirces anti-Cartesianism was John Dewey. He combined
it with Jamess emphasis on the acquisition of knowledge as an active and exploratory
process, rather than a kind of passive contemplation, with the view, present in much
pragmatism, but particularly emphasized by Schiller, that the conceptual instruments of
thought are human constructions not independent Platonic existences and, third, with an
insistence on the social character of knowledge that is to be found in Peirce, to produce
a distinctive theory of knowledge that has a remarkable coherence of tone despite the
breadth of its scope and, it must be admitted, the frequent turgidness and amorphousness
of its expression.
It is this theory of knowledge of Deweys that I shall be discussing today. I find all of
its main contentions at least suggestive even if, in the form in which Dewey presents them,
they are open to much criticism. In general, those constituents of the whole for which
Dewey is himself chiefly responsible are, I shall argue, more acceptable as regards what
they deny than as regards what they positively affirm. It is still odd that a body of extensive
critical discussion that addresses itself to such fundamental aspects of the traditional British theory of knowledge should have secured so little attention here.
2 The four themes of anti-intellectualism

A term is needed to pick out Deweys theory of knowledge from pragmatism in general.
His own preferred designation, instrumentalism, is not altogether satisfactory since it concentrates too much on one part of the whole. It is perhaps best characterized in negative
terms as anti-intellectualism. It can be set out by listing its four main points of conflict and
disagreement with the familiar Cartesian tradition.
1 In the first place where the Cartesian seeks to base all knowledge on absolutely certain beliefs, on somehow self-evident items of intuitive knowledge, the anti-intellectualist
contends that all our beliefs are fallible and corrigible. In consequence the task of epistemology is not to give an account of secure and certified knowledge but rather of rational
and warranted belief. In general, the beliefs on which we act are not established certainties.
The demand for established certainty is exaggerated and utopian. We must be content with
warranted assertibility which falls short of absolute and ideal truth. It is in the spirit of this
conviction that Dewey puts himself forward as a theorist of inquiry rather than a theorist
of knowledge.
2 Second, the anti-intellectualist sees the knower or inquirer, the pursuer of rational
and warranted belief, as an active being, an experimenter, not as a contemplative theorist.
It is in this connection that Dewey launches his polemics against the spectator theory of
knowledge. Our rational beliefs about the world in which we live and act are not the result
of a kind of Augustinian illumination, passively received, and then privately worked up and
Inquiry, thought and action: John Deweys theory of knowledge 3

systematized in the recesses of our own minds. They are, rather, the outcome of deliberate, experimental interaction with our environment. Furthermore the intellectual apparatus
which we bring to this cognitive encounter with the world is not something imposed on
us by the pervading structure of the world outside. Our conceptual equipment is a body of
instruments that we have devised and constructed ourselves, under the pressure of our own
needs and purposes.
3 Third, the subject or possessor of knowledge or warranted belief is not, in the manner
of intellectualism, a pure mind or consciousness, a Cartesian res cogitans. It is, in Deweys
view, an intelligent organism, an embodied thing, animated by primarily bodily purposes,
and forming its beliefs about the world around it through bodily, physical interaction with
it. Manipulation is at least as crucial to the formation of rational beliefs as more or less
detached inspection.
4 Fourth and finally, the anti-intellectualist sees the pursuit of rational belief as an essentially social undertaking, in contrast to the subjective isolation of the Cartesian knower,
exposed as he is to various kinds of sceptical desperation. Knowledge or rational belief is a
social product, an accumulation of common intellectual property, made up of what Dewey
likes to call funded experience and on which all may draw.
In short, the intellectualist sees knowledge as something absolutely certain, which is
contemplatively seen, by a mind that is at most contingently embodied, working on its
own. For Deweys anti-intellectualism what is sought is rational and corrigibly fallible
belief, actively achieved, even made or constructed, and with the aid of conceptual instruments of human design, by an intelligent but embodied organism that is a natural part of the
world it seeks to know, engaged on this undertaking as a collaborating member of a society
of intelligent organisms of the same kind.
3 Fallibilism
Let us consider the first theme, that of fallibilism. Intellectualism, particularly in its Cartesian and intuitionist form, defines knowledge as conclusively justified true belief and maintains that all rational belief must rest on knowledge thus defined. The anti-intellectualist
is opposed to the suggested definition of knowledge as utopian and even superstitious
but does not so much attack it by direct argument as simply replace it by a conception of
knowledge held to be more realistic and useful.
In effect, that is to say that we should not bother ourselves with absolute certainties
since they are not to be had and, fortunately, are not needed. What, he asks, is knowledge
for? Primarily, persistingly and essentially for the sake of action. What we require is rational belief about the consequences of various alternative actions it is within our power to
take. In being essentially practical, it is also essentially forward-looking and predictive and
all such prediction is fallible.
In a way there is nothing here with which Cartesians generally, and Cartesians of an
empiricist variety in particular, need quarrel. If absolute certainty is insisted on as a defining condition of knowledge then it will turn out that very little is known. Specifically what
is truly known is whatever is intuitively necessary, whatever is in some way empirically
self-evident and incorrigible, in other words the immediate deliverances of sensation and
introspective self-consciousness, and whatever can be deduced from premises of either of

these sorts by means of rules of inference corresponding to intuitive or demonstrated necessary truths. On this view most of what is of practical interest for man as an active being
will not be known. Cartesian empiricists will agree that with regard to the material world,
the past, the minds of others and the laws of nature, we can have at best rational belief, for
all propositions about these are neither intuitive nor demonstrable. For the most part they
will be the conclusions of non-deductive inference which are at best confirmed and never
certified by the evidence on which they rest. (Knowledge of the past by memory is not
inferential, of course, but it is as fallible as if it were and thus as little entitled to description
as knowledge proper.) No one could be more emphatic that the beliefs of science and common sense are not truly knowledge than Russell, unless it were Hume.
But, if the Cartesian tends to admit that very little of what we believe, and, in particular,
practically none of the beliefs which directly guide our actions, are truly known in his sense
of the word, he still insists that the little that is truly known is of the utmost importance.
This is because of his second thesis that there can be no rational belief that is not derived
from knowledge proper.
Dewey does not, so far as I know, attack this thesis directly but I believe it to be mistaken. It is commonly presented as the only alternative to a coherence theory of knowledge
in which every belief owes its justification to its inferential relation to other beliefs (a position something very like that to which Peirce was driven by his rejection of intuitionism).
This view leads to an infinite regress which presents every appearance of being vicious.
Beliefs cannot simply pass justification from one to another without, so to speak, their having some initial stock of justification from another source.
But such a coherence theory is not the only alternative to intuitionism. The fact that all
beliefs that derive whatever justification they have to non-deductive inference from other
beliefs are less than absolutely certain does not entail that all beliefs that are less than certain are inferential in the way described. In other words it is not clear that all rational, but
less than absolutely certain, belief must be derived from beliefs that are absolutely certain
and incorrigible.
It is possible to hold that there are foundations to knowledge, or, more precisely, relative
foundations, which, while not absolutely certain, are nevertheless rationally believed, not
on the strength of other beliefs that inferentially support them but on the strength of experiences that do not certify them but simply render them more worthy of acceptance than
rejection. Just this is true, I believe, of the propositions about perceived material things and
the remembered past that are the apparent, rather than theoretically alleged, foundations of
our empirical beliefs about matters of fact.
If this view is accepted it is not necessary to suppose that our beliefs about the material
world as we commonly take ourselves to perceive it are really inferences from antecedent
beliefs about our sense-experiences or again that our beliefs about the recollected past are
inferred from antecedent beliefs about memory-data.
Although these basic empirical beliefs about the perceived material world and the recollected past are not initiated or wholly dependent for their justification on inference from
other, more certain beliefs, they are, nevertheless, subject to inference, in so far as they can
be further confirmed by other beliefs of the same corrigible but credible status, in the light
of equally corrigible general beliefs about the way in which the constituents of the material

world behave and hang together. Equally, of course, they may, despite their initial credibility, be undermined by inference from a preponderating body of comparable beliefs.
This brings us back to the definition of knowledge in terms of absolute certainty. What
this defining condition in fact comes to is that a proposition can be known only if it would
be an evident or demonstrable contradiction to deny it or if it follows from the fact that it is
believed that it is true. In either of those circumstances there would be some kind of logical
absurdity in the supposition that the belief in question might be false. But without going to
the length of blithely re-defining knowledge as well-confirmed belief, it might be felt that
that is altogether too stringent an account of certainty.
The certain, it could be suggested, is not that which it is somehow logically absurd to
doubt, but that which there is no reason to doubt; that which rationally ought not to be
doubted, not that which logically cannot be. Now in these terms although no basic empirical proposition may be certain as prompted by experience alone, it may acquire certainty
from the addition to that confirmation of the support of other equally fallible beliefs. Certainty, on such a view, is something ascertained, not something initially given. But it could
be what we ordinarily understand by the word.
I conclude, then, that the Cartesian does define knowledge too restrictively and that he
is mistaken in thinking that its uninferred empirical basis must be incorrigibly certain. But
such ideal certainty is not really essential to knowledge and ordinary certainty is attainable
for basic empirical beliefs even if they are not endowed with it by the experiences that
directly prompt them. In rejecting the Cartesians definition of knowledge Dewey goes
too far in the opposite direction and, in his emphasis on the fallibility of most practically
important beliefs, says nothing that an empirically minded Cartesian could not accept,
while failing to engage with their real mistake: the assumption that the foundations of
knowledge must be logically immune from error.
4 Instrumentalism
The focus of Deweys epistemology is his attack on the spectator theory of knowledge
on behalf of instrumentalism or experimentalism. At a certain level of generality the idea
that knowledge is actively and purposively sought, not just passively received, is likely
to secure acceptance without much difficulty. It might indeed be said that our acquisition of knowledge or rational belief is neither wholly active nor wholly passive. Surely a
certain amount of what we know or reasonably believe just gets borne in on us and does
not take the form of answers methodically secured to antecedently formulated questions.
The process by which I have come to know that I do not like the taste of liquorice, that my
colleague is in a rather irritable state today or that a friends wife has put on a good deal of
weight, does not deserve to be dignified with the description inquiry.
In very general terms like these there is plainly much to be said on behalf of an active
attitude in the knowledge-gathering operation. In the tone of an epistemological Polonius
one could urge that if knowledge is actively sought it will be both better founded and a
great deal more copious than if it is merely allowed to accumulate in a passive manner. It
is this active attitude, after all, that is the most fundamental mark of distinction between
science and common sense; for whatever else it may be, science is at least a deliberate and
methodical effort to answer questions about the nature and connection of things. If beliefs

are just allowed to form by a kind of natural accretion they are perhaps peculiarly likely to
be incorrect. To produce a final bromide: one generally does something better if one attends
to what one is doing.
At the same level of rather bland generality is Deweys insistence that the process of
inquiry is prompted and set in motion by mens practical needs, that knowledge is for the
sake of action. Here again, one may readily admit that a great deal of knowledge is of this
kind, perhaps most of the knowledge that most people have. But there is such a thing as
pure curiosity, as distinct from the curiosity which arises from a suspicion that something
may have hitherto unknown properties that call for action, perhaps of a pre-emptive kind,
and, one has to add, from the impure curiosity of the inquirer who has to find something out
in order to pass an examination or retain a job, however practically indifferent he may be
to the content of what he is seeking to find out. Polonius is waiting in the wings once again
at this point, bursting with the information that many things that people have been motivated to discover by the love of knowledge for its own sake have later turned out to be of
the highest significance for practical human purposes. One may sympathize with Deweys
anxiety to make philosophy serviceable to men in general in the concerns of everyday life
without feeling obliged to make out its credentials in this respect in all its aspects.
Instrumentalism takes on a more concrete and definite character in the form of the thesis
that the materials of belief, the concepts in which beliefs are formulated, are human constructions and not imposed on men by the nature of things. This thesis is directed against
intellectualism of a Platonic kind which takes the conceptual materials of our thought to be
somehow imposed on us by the nature of things. Our conceptual apparatus, on this view,
reflects, to the extent that it is adequate for its task, the structure of an objective and timeless realm of essence. Dewey was always hostile to Platonism on more or less democratic
or egalitarian grounds. He took it to be the attitude to knowledge appropriate to a slaveowning society in which true rational men or citizens did not soil their hands with the work
of the world but sequestered themselves for purposes of abstract, theoretical contemplation. But there is a certain vulgarity about this opinion. Mathematics and metaphysics are
not dirty work, except in the marginal forms of computer engineering and sorcery, but that
does not mean that they are not work, that they are not fields of active, answer-seeking
effort, typically, I should suppose, more so than the routine discharge of practical tasks.
The active, imaginative invention or construction of concepts is most evident at the
level of scientific theorizing. The theorists intellectual fertility does not show itself only
in bringing familiar concepts together in previously unformulated beliefs. It is also present in the devising of new concepts that are not part of the common stock of thought and
discourse: elasticity of demand, deep grammatical structure, the quantum of energy, the
correlation coefficient.
Concepts of this kind differ from more familiar ones like red or square or tree or dog in
that they have a history, a known history, that is to say, providing a date at which they were
first introduced and the name of their first introducer. But two points need to be made here.
The fact that a new concept was introduced into discourse at a particular time by a particular person does not mean that it is strictly an invention. The conceptual innovator may
be just as well described as having brought off a feat of discovery. The actual or possible
property or relationship that he succeeds in bringing to human consciousness may well
have already been exemplified in the world, even if no one before him was aware of the

fact. Indeed, if the new concept is to be of any use in the formulation of true beliefs, what
the concept expresses must, in most cases, have been, or be going to be, actually exemplified. (The point of the qualification is to allow for concepts of ideal or limiting cases: the
perfectly elastic fluid, the ideal gas, the economy in perfect equilibrium.) The conceptual
innovator, in other words, must have one foot on the earth.
Second, although the fact of historic innovation may show some concepts to be the outcome of acts of imaginative creation, those concepts will be theoretical and sophisticated
ones like the examples I gave. The ordinary notions with which they are contrasted, the
everyday descriptive apparatus, would seem to be a piece of common human property of
unhistoric antiquity and unhistoric immunity to reform and change. One rather obvious
piece of evidence for this view is the translatability of ancient or geographically remote
languages into the language we speak ourselves.
It is, of course, true that there are limits to translation. I am thinking here not of the
problems raised by Quine but of the problem posed by Homers wine-dark sea. Even the
most elementary and familiar concepts have a sort of history and we may reasonably suppose them to have emerged as part of the natural evolutionary development of mankind.
That point of view undermines the old idea of a universal and identical human reason to
be found in Aristotle and the Stoics and defended by Kant on the ground of the universal
validity of logic. It seems reasonable to suppose that our conceptual apparatus is not a
direct reflection of the nature and structure of the world but the result of an interaction,
worked out in an evolutionary way, between the world and two things located in us: our
perceptual equipment, on the one hand, and our needs and interests on the other.
The view of Platonic intellectualism, that our conceptual apparatus directly reflects the
structure of the world, does not have as its only alternative the idea that our conceptual
apparatus is a wholly free construction, imposed by us on the indefinitely plastic tissue of
our environment. Between biscuits and clear homogeneous soup is minestrone. We can
allow ourselves. that is to say, our perceptual equipment and our needs and interests, an
important measure of free initiative in the formation of concepts without supposing ourselves to be absolutely free in this matter. The initiative in question is to select, from all the
possibilities of comparison or similarity-finding that are present in the world, those that our
perceptual equipment enables us to register, those which present themselves to the close
attention that is excited by need and interest and those, finally, which while not thrusting
themselves on perception nor directly ministering to an interest, allow for the conveniently
brief formulation of laws or are, as one might say, explanatorily fertile.
The mind or knower, then, can be admitted to be conceptually creative without denying that the conceptual outfit it creates is some kind of reflection of the world; for surely,
if it did not in some way reflect the similarities and differences to be found in the world it
would be descriptively useless. If our thinking is to be communicable from one person to
another the words in which they express it must, where the same, apply to much the same
things. We must share dispositions to classify things together and to distinguish them. If
that is to happen things classified together must strike us each in much the same way and
things distinguished must strike us differently, and for that to happen the things in question
must actually be, respectively, alike and different. I conclude that the valid part of Deweys
instrumentalist theory that concepts are human constructions is that facts about human

beings determine the selection of those features of reality that are conceptually registered
but that it is genuine features of reality that the selection is made from.
Deweys critique of the spectator theory of knowledge does not confine itself to the
passively intellectualist account that theory gives of concept-formation. The formation of
beliefs, Dewey maintains, as well as that of the material of belief, is a species of human
action, not just the passive absorption of what he calls antecedent reality. One slightly
puzzling way in which this position is expressed is in the statement that a belief is a plan
of action. It is clear enough that beliefs are frequently parts of plans of action, in the ordinary sense of the phrase; indeed, it is hard to think of anything that could be called a plan
of action that does not at least imply or presuppose some belief or beliefs. But a belief can
plainly be held without being part of any actually formulated plan of action in the mind of
the believer, such as my belief that Sirius is a very large star.
Dewey generally sees inquiry as the result of some obstacle to action and no doubt it
often is. In this situation the inquiry succeeds in removing the obstacle to action by arriving
at a belief which makes the formation of an effective plan of action possible. The door is
locked and I cannot get in. I look inquiringly for the key under the mat, find it there, and,
in a twinkling, I form and put into effect the plan of picking it up, putting it in the keyhole
and opening the door.
The process of inquiry that terminates, if successful, in the formation of a belief will
typically be itself a form of action. In the minimal case the action is that of just looking
attentively. But commonly that is not enough. I see something that looks like a, key in the
shadows above the lintel. I think to myself if that is a key it will feel hard and cool to the
touch and put into operation the verificatory manoeuvre of reaching out to touch it. The
confirmation of initially insecure beliefs is an exploratory, experimental process of action,
so the belief-forming process of inquiry is itself a kind of action. But it is not, so to speak,
all action. The hypothetical statements that set out potential confirmations of what we are
inclined to believe suggest experimental lines of action that we can follow. But once the
action suggested has been performed we just have to wait and see (or feel) to exercise our
negative capabilities of sentience. My reaching out and touching the key is something I do;
the hardness and coolness I then feel it to have is something the world (or the key, if that
is what it is) does to me.
It is through an extreme of nebulosity at this point in his account of the nature of inquiry
that Dewey arrives at his most surprising conclusion: inquiry is the controlled or directed
transformation, he writes, of an indeterminate situation into one that is so determinate in
its constituent distinctions and relations as to convert the elements of the original situation
into a unified whole. Or again: the outcome of the directed activity [sc. of inquiry] is the
construction of a new empirical situation in which objects are differently related to one
another and such that the consequences of directed operations form the objects that have
the property of being known.
In any situation of inquiry three sorts of change may be brought about. To start with
there is change simply on the side of the inquirer. He was in a state of ignorance (I dont
know where the key is); he arrives at a state of knowledge or well-founded belief (Here
is the key). Second, there may be some change introduced by the experimental activity of
the inquirer which affects the object of his inquiry. He picks the key-like thing up, has a
good look at it, and concludes that it is a key. Here the key is no longer where it was, but in

his hand and near his eyes, but it is still a key. Finally, he may bring about a change in the
object which, as it were, frustrates the original purpose of the inquiry: Where is the key?
Well, its in my hand now. Or, more catastrophically, Where is the egg? It was in my
hand but now its all over the place.
The third kind of change is not the essence of effective, practical inquiry; it is its nemesis. It is something to be carefully guarded against, not welcomed. We murder to dissect
is not a metaphysical truth, but it is a methodological caution.
Beliefs are, then, often for action and they figure essentially in plans of action. In seeking to arrive at well-founded beliefs we commonly engage in experimental activity which
is intended to produce a change in us and is ordinarily going to produce some change, even
if not a very central or intrinsic one, in the object experimentally acted upon, its relative
position, for example, its condition of illumination and so on. But if it centrally alters the
object itself the experimental aspect of inquiry defeats its own purpose. At some stage in
inquiry the inquirer must be a spectator, however questioning and actively experimental or
manipulative he may be at other stages. The object must be left room to do its part. If we
put it to the question, it still has to give the answer. Dewey seems to represent the inquirer
as a kind of inefficient torturer who does not discover the crucial thing his victim knows
but rather what his victim thinks he would like to hear. Dewey is right to stress that inquiry,
the acquisition of knowledge or rational belief, is an interaction between the object and its
investigator. But it is an interaction, not just the converse of the one-way process presented
by the spectator theory of knowledge.
5 Naturalism
Deweys emphasis on the active nature of inquiry rather strongly implies that the inquirer
should be conceived as an intelligent organism, physically interacting with the objects he
investigates, and not just as a contemplating mind. It does not strictly entail that conclusion, since a disembodied Cartesian mind could be thought of as being in a way active in
his contemplative operations, if only by the directing and focusing of his attention. But
Dewey would rightly insist that in fact the confirmation of our beliefs involves bodily
manipulation of their objects.
Many philosophers have pointed out the close connection between the spectator theory
of knowledge and the tendency of theorists of knowledge to identify perception with sight.
In visual perception, for the most part, the action of the body is minimal: sometimes it is
just a matter of focusing the eyes; on many occasions no more bodily action is required
than moving the head.
But touch, after all, is as important to our perception of the external world as sight. As
Warnock has said, although sight is very informative and detailed in its deliverances, we
rely on touch in the end for a final check on what sight prompts us to believe.
No one, in fact, would deny that the perceiver is, at least in part, a physical thing causally interacting with the rest of the physical world, or that causal influence exercised on the
physical mechanism of perception by its objects is a necessary condition of anything being
perceived at all. But what Cartesian theorists of knowledge would maintain is that these
propositions, although true, do not affect the epistemology of perception. The knowledge
they express, although it is perfectly genuine knowledge, for which any adequate theory of

knowledge must find a place, is still knowledge of a secondary, derivative, theoretical kind.
They can invoke the same arguments they have used to show that in visual perception all
that we really or directly perceive is private entities in our own streams of consciousness
to domesticate or subjectivize the deliverances of touch and organic sensation. There are,
after all, touch-illusions, such as the effect of rubbing ones hands up and down together
with a wire mesh between them, which makes the wire feel like cloth or silk, and once these
are admitted there is an entering wedge through which the distinction between the touchdata we directly perceive and the actual tactile properties of things can enter.
In a thorough discussion of these neglected areas of touch and organic sensation
H.H.Price interestingly traces the conception of causation as some kind of active efficacy,
and not just the kind of regularity which is all that vision finds in it, to the experience of
forceful resistance we have, by way of muscular sensation, when objects impinge on us or
we press against them. At one point he says that the resistance we experience is essentially
relational in nature and so no inference is required to establish an external resistor. But he is
equally emphatic that we must not fail to distinguish the sense of embodiment which is the
constant background to all our sensory experience from the fact of being embodied which
that sense no doubt encourages us to accept as a fact but is not entailed by it.
To insist that we are, as perceivers, embodied organisms in physical interaction with the
external world we perceive does not really undermine the sceptically Cartesian account of
the indubitable foundations of empirical knowledge. It only highlights in a forceful way
the oddity of the Cartesian account of the knower. That oddity is not enough on its own
to refute the Cartesian position. After all, the Cartesian is usually going to find a place for
these facts in his overall view of the structure of knowledge. To refute his basic subjectivism it is necessary to confront it more directly, as is done by the defence of fallibilism that
I expounded earlier. But once we free ourselves of the ultimate Cartesian principle that all
our knowledge begins and owes its ultimate confirmation to facts we perceive immediately
about the contents of our own minds, the fact that we are, as perceivers, embodied organisms, physically interacting in perception with the world that we perceive, can be placed at
the logical and psychological beginnings of our acquisition of knowledge about matters of
empirical fact and not be represented, as in Cartesian subjectivism, as a matter of more or
less sophisticated and precarious theory.
6 The social nature of knowledge

Every theorist of knowledge would admit that most of what we actually know or rationally
believe we owe in some way to others. In recognition of this fact they usually append to
their lists of the sources of knowledge some reference to testimony or authority. It will be
tacked on in a rather undignified way at the tail end of a sequence that begins with perception and runs through self-consciousness or introspection and memory to inference. Quite
often, indeed, testimony is regarded as a special case of inference, deriving its conclusions from premises of the form A says p and most of what A says is true. The first of
these is established by perception, the second by induction from observed correspondence
between what A has been heard to say in the past and what we have found out to be true
on our own.

There is an interesting problem about testimony. Plainly we believe a lot of it when we
are not in a position to affirm the generalizations about reliability which would be needed
for such beliefs to be rationally accepted. What is more, when we do come to check on the
reliability of external informants, we do so with critical instruments with which we have
been externally supplied. I have argued elsewhere that this problem can be solved, that
the tests we use on testimony are not, as the problem suggests, potentially corrupted at the
source. The nerve of the argument is that other people could not mislead us about logic and
perception, which are all the testing instruments required. They could perhaps prevent us
from learning to speak at all, by energetically random utterance in our presence, but if they
are to teach us to speak at all they cannot help teaching us more or less correctly, that is to
say in accordance with the rules with which their observable practice has conformed.
Cartesian minds are isolated things, epistemological Adams or Crusoes, making their
way in the world on their own. Deweys intelligent organisms pursue warranted beliefs in
a society of other inquirers like themselves and in communication with them. This, once
again, is something we all know but it occupies a very small and marginal place in Cartesian theories of knowledge. To draw attention it is not, as it stands, an argument against
such theories, except to the extent that it brings out the extent to which their assumptions
about the foundations of knowledge lead them largely to ignore the actual character of
knowledge in a world of social beings. Deweys pursuit of a theory of knowledge that will
be concerned with the actual problems of men here once again opens up a range of problems about knowledge which the tradition to which he is opposed has largely neglected.
2
Language and experience
Jerome Bruner, Eileen Caudill and Anat Ninio
Let us begin with a Dewey theme. It is that the shape and structure of human experience
and human action are reflected in the very nature of language, that language is not itself
a system of logic, and that more precisely, the uses to which language is put by any given
individual, the linguistic procedures he will employ, necessarily reflect the circumstances
in which he has lived and how he has coped with them. In a word, language itself is in
some deep sense a record of human experience and its particular personal manifestation is
a record of individual experience. In contemporary jargon, language is never to be understood as context independent.
This theme would surely be neither surprising nor particularly timely were it not for two
historical circumstances. The first of these is the revolution that has occurred in the study of
language over the past two decadesparticularly since 1957 when Noam Chomskys Syntactic Structures exploded like a star shell in the world of linguistics. Now, better equipped
with data and with manageable doubts, we are able to re-evaluate what that revolution
accomplished and what it distorted. Alfred North Whitehead, saluting Lord Russell for a
talk he had just given at Harvard on the meaning of quantum theory for the philosophy of
science, ended by thanking him for a lucid account that had managed not to obscure the
great darkness of his subject. I think we can say of the Chomskyan revolution in linguistics
that while it elucidated the nature of syntax, it did in fact obscure the darkness of other, perhaps more important, aspects of languagethose precisely that were context dependent.
No doubt, by sharply distinguishing the structural from the context-dependent features
of language, Chomsky moved formal linguistic theory ahead. My only worry is that that
forward movement was dearly paid for by those interested in the acquisition of language,
where the separation leads to confusion. But I am not a linguist. My concern over the last
several years as a psychologist has been to understand how it is that the child so quickly,
and seemingly so easily, learns to use the instrument of language. So swift is the course
of that learning that we have been tempted to assume that the child has an innate capacity
for language and that this capacity existed in the form of innate ideas about grammar. That
conclusion was forced artificially by Chomskys sharp distinction.
This brings me to my second historical circumstance. It is the new turning point that has
been reached in the study of language acquisition, and since history requires an apparatus
of dates and places to make it memorable, I shall choose as my locus London, the occasion
being the Third International Child Language Symposium held in September 1975 at the
School of Oriental and African Studies. Two things were evident at that congress. The first
was that the doctrine of syntactic primacy in the study of language had died, perhaps of
fatigue, and the second was that a new and interesting period had begun in which experi-
Language and experience 13

ence and function had emerged afresh as central to our understanding of what makes it
possible for the child to pass so quickly, and so seemingly effortlessly, into the initial use
of language. With this new realization, the Dewey theme with which I started this lecture
returns to the centre of the stage. I should like to use the occasion of this talk to re-examine
it in the light of work now in progress on language acquisitionsome of it, indeed, being
conducted by my colleagues and myself at Oxford, but much of it scattered from Stanford
to Warsaw, from Edinburgh to Jerusalem.
Before I start, first with the demise of syntactic primacy and then with work on language
acquisition, let me say a few words about John Deweys view of language and experience. You will recall the central point in his discussion of the relationship of thinking and
experience in his celebrated essay on Thinking in education. He comments: No thought,
no idea, can possibly be conveyed as an idea from one person to another. The communication may stimulate the other person to realize the question for himself and to think out
a like idea. In his discussion of this point he makes much of the relationship between
experience and thought. Experience for Dewey involved two conjugate components: an
active and a passive one, which he characterized by the plain terms trying and undergoing. Meaning emerges, thought and experience are connected, when trying and undergoing operate jointlywhen the former, trying, develops the connections and consequences
inherent in undergoing. What makes difficult the transmission of ideas by communication,
in Deweys view, is precisely that the process of connection cannot be achieved by the
communication itself, but only by the activity of the recipient goaded by the communication, aided of course by the possibility of transaction between members of a dialogue in
the interest of making ideas clearer. In fact, Dewey was little concerned with the details
of language, and probably his adherence to particular doctrines, like the pragmatic theory
of truth, would be found wanting by thoughtful philosophers of language. But what is
extraordinarily clear in his writing is a view about language comprehension as an active,
constructional processtrying rather than undergoing, in his sensethat may serve us
later in examining the state of our knowledge about language and experience.
So let us turn to the doctrine of syntactic primacy in language acquisition, its rise and
fall. It begins with Chomskys efforts to describe formal discovery procedure as one of the
goals of linguistics. Such procedures could be seen as mechanisms that take the corpus of a
language (or some sample thereof) as input, yielding a grammar as output. The first actual
proposal for such a machine was made in a conference paper by Miller and Chomsky in
1957, but that paper was never published and the two authors have since lost the copies
they had. So the birth of LAD, like most heroic births, is wrapped in the mystery of the lost
child. The idea that such a machine might serve as the basis for initial language acquisition
by the child seems not to have occurred to Chomsky until 1960 (see Chomsky, 1962) when
he set forth the following proposition at a conference:
We might attempt to construct a device of the kind:
(1) utterances of L
formalized grammar of L.
This represents a function that maps a set of observed utterances into the formalized
grammar of the language of which they are a sample. Given as input a sufficiently large and
representative set of utterances of any language (English, Chinese, or whatever) the device
(1) would provide as output a formalized grammar of this language. A description of this

device would therefore represent a hypothesis about the innate intellectual equipment that
a child brings to bear in language learning.
The device was later to be called the Language Acquisition Device or LAD.
It had several notable properties, and since this account is meant as an obituary rather
than an analysis, we shall only single out those that have some relevance to our future
consideration of the deceased. For one, it was postulated that mastery of the grammar of a
language was independent of mastery or experience outside the domain of language. LAD
was thought of as a hypothesis generator, trying out hypotheses based on innate linguistic
universals for their fit to the surface structure of the local language. Other forms of experience were irrelevant. A second feature, of course, was that meaning had little or no place
in the system of acquisition, nor was language acquisition affected by the extra-linguistic
function for which the language was being used. This was consonant with the third feature
of the theory: the input corpus of the language was in effect an overheard sample of speech
and no specification was made concerning how that input had to be regulated. The learner
was, so to speak, an eavesdropper at the adult language feast.
That is sufficient, although it may leave some of you not familiar with the details,
rather in a cloud with respect to such terms as linguistic universal, surface structure, and
so forth.
The only point I wish to make is that the doctrine of syntactic primacy rested on the
idea that language was separate from experience, that its acquisition was based on innate
appreciation of structure without regard to meaning and use, and that how the language
was encountered did not matter, save that the encounters provide a decent sample of the
surface structure of the local language that would allow for recognition of its likeness
to linguistic universals. I think I can say, without either the danger of exaggeration or of
denigrating the enormous contribution of Chomsky, that each one of these conceptions
was deeply, non-trivially mistaken. Language acquisition does depend upon mastery of
non-linguistic concepts (although it is not explained by them); the mastery of syntax is
dependent upon meaning and it may well be better to think of initial syntax as itself serving
a semantic function; and finally, the corpus of speech that is the input upon which acquisition is based is not a sample, not overheard, not indifferent. In fact, the corpus of speech
to which the child is exposed is governed by rules of interdependent interaction between
child and adult and the uses to which language is put are powerfully important in how the
child gets into the language. Indeed, in some respects, ironically enough, the adult could be
thought of as a kibitzing eavesdropper at the childs linguistic feast. Or better yet, the child
and the mother-tutor learn how to use each other as linguistic informants. There may be
one additional point worth making, for it relates to the issue of generativeness in language.
As Ferrier commented at the International Symposium and as we have also noted, the child
is exposed to an extraordinarily limited and repetitive range of learning during the period
when he is first getting started on his career as a speaker. The same things are said over and
over again in like contexts.
But please note that understanding the world by forming good working concepts of it,
depending upon meaning to get the hang of what is being said and how it must be said,
and having continual interaction between mother and child as a dialogic carrier-wave for
language, still does not account for how the child learns the lexico-grammatical procedures

of language, or how he swiftly masters its subtle and systematic phonology; for the procedures of language, while not arbitrary in my view, are not of the same stuff as the concepts
by which we organize the world to which language refers and which language is capable
of representing. The syntax of a language, the relation of words to each other in utterances,
still awaits acquisition: its category rules, privileged orders, its techniques of inflecting, its
pre- and post-positioning, its modes, its methods of tense-marking. These are not explained
away by saying that LAD is insufficient or even that it is a brave but silly idea, a logicians
fantasy. We are now, rather, face to face with the task of trying to understand how the
former set of thingsthe childs knowledge of the world, his knowledge of interaction,
and his sense of the extra-linguistic functions that must be served by procedural rules of
languagehow this knowledge helps him master the latter set of thingsthe actual procedural rules that make up the linguistic code. Simply because the effort to explain it all by
assuming that mastery of syntax is innately determined has failed, does not mean that we
have now succeeded by other means. The work still lies ahead, and it is that work to which
we must now turn.
I should like to organize the discussion around two general problems. The first has to do
with what Michael Halliday calls the mathetic function of languageinforming, getting
knowledge, using heuristic devices like questions, etc.and I want to confine myself to
that range of events that starts toward the end of the first year with pure pointing and ends
with the astonishing process of labelling and primitive predication. When last I spoke at
this university, about a year ago delivering the Doris Lee Lecture, I was concerned with the
period before that. We go on from there this evening. The second topic is, to use another
Hallidayan term, the pragmatic function: how the child uses language performatively (as
John Austin would have put it) to negotiate his relations with othersgetting them to help
him, expressing his affiliation with them, regulating their behaviour.
About the mathetic function, in my previous lecture of last year, I spoke of the manner
in which mother and infant came to manage joint reference by procedures for singling out
which among a set of alternative events or objects they each had in mindby monitoring
each others gaze direction, by using distinctive phonological markings, by developing
primitive and idiosyncratic labelling procedures, by developing demonstrative heuristics
like showing, etc. Most of this activity had the property that what was attended to by the
child and what he brought to his mothers attention tended to be in the category of events
that might be called objects of desire. Attention, in the main, subserved intended action
with respect to objects. The exceptions in large part were certain familiar persons, Mama,
Dadda, baby-sitter, etc. They were the recipients of greetings, of inexplicable labelling, etc.
So, too, certain familiar objects that achieved special labelling.
With pure pointing, something else enters the scene. Pure is used to characterize the
kind of pointing that is not an abortive reach toward a desired object. It is a distinctive gesture, involving an extended forefinger, and in the child, Richard, about whom most of my
remarks are relevant, it begins at about 12 monthsperhaps a little late by the usual norms
we find. Before I comment on the contexts in which such pointing first appeared, I should
say something about a classic issue in linguistics that I am sure many of you are acquainted
with. It is the issue of the familiar versus the unfamiliar, that which is explicitly marked
by utterance or gesture or other procedures, and that which remains unstated, presupposed,

implicit. We have some indication that before pointing occurred, Richard was already making the distinction by the use of some idiosyncratic vocalizationsgreeting familiar situations and objects and pictures with either geki or dede. It is not the least surprising that
he could distinguish the two kinds of situations or that he found ways of indicating that
he couldfor example, he would look back at his mother when faced with a novel object,
not with a familiar one, or smile at familiar situations. What is surprising is that he should
embody the distinction in a vocal gesture.
The first steady rounds of pointing occurred when Richard went on holiday with his
family to the Lake District. His date for videotaping occurred during this absence, so we
followed him: his age 14 months 3 weeks. (There had been 6 pure points in a half-hour session when Richard was 13 months 1 week.) The filming was done out of doors, in relaxed
rural surroundings, much of it novel to Richard. During the 30 minute period of observation put on tape, there were 35 pure points. In a 3 hour period of baby-sitting plus an hours
walk, more than a hundred episodes of pure pointing were observed, the type of high-frequency occurrence that characterizes new functions early on. All of these were directed to
objects that had the following characteristics: (a) if objects were distant more than a metre
and therefore out of reach, (b) they were, as objects, in some way unfamiliar or unexpected,
(c) they were, if neither an object nor novel, a picture or drawing of a familiar object, or (d)
if they were none of these, they were imaginary. By an object is meant person, animal,
or thing contained within a small and finite locus. It will immediately be apparent that the
rule is one that might be called a distancing principle: pointing occurs when an object is
just distant enough, in some dimension, to be interesting and yet not an immediate object
of desire to be grasped or pursued or possessed. It may well be that, indeed, some of the
objectscows and sheepwere both interesting and possibly a bit frightening. Let me
illustrate with instances of the rules. Distant, unfamiliar objects: sheep, cow, baby in pram,
bird, experimenter. Depicted familiar objects: at least six familiar objects in a picture book.
Imaginary objects: pointing upward and saying /b/, the designation also when pointing at
birds seen flying above him. Most instances of pointing were accompanied by vocalizations, usually of a non-specific, non-standard type. The example of /b/ is an exception.
Now, there is one big exception to the rules, and it is an exception produced by the
nature of dialogue between mother and child. The mother can be conceived of as having the special role of rendering the novel and the unfamiliar quite banal in response to
the childs pointing. The first recorded instance of pure pointing entered in the diary his
mother keeps for us (12 months 3 weeks) is directed toward a woman visitor, a stranger to
Richard, at some distance from him. This is a sign for the mother to label with the usual,
Yes, Richard, this is Mrs So-and-so. This illustrates one of the ways in which pointing is
handled: by labelling. The other is by the mother bringing the distant object indicated by
the childs pure pointing into the childs reach or possession. At 14 months 3 weeks we
have recorded a session in which Richard points to a flower. His mother labels it, picks
the flower, and gives it to Richard. He runs his extended forefinger around the edge of
the blossoms, silently, and very quickly discards the flower. His attention is not caught by
the possibility of manipulation. It would seem that pointing is a technique for marking an
object, person, or event, to proclaim its presence to another. When he points to an object,
he either then looks at his mother or she is so close by (as in book-reading) that her joint
line of regard is presupposed. It is of note, too, that he almost always vocalizes in some

way. (Another of our children, Jonathan, was observed upon waking and before his parents
joined him, to point to a picture on the wall, but without vocalization, the video camera
having been set up the night before. But generally, his pointing starting at 10 months was
accompanied by /Eh/ when others were present.)
An example of the marking use of pointing for dealing with the novel is provided by
Jonathan at 13 months 1 week. We brought a set of new toy animals to a session at his
house and placed them on the mantel out of reach. Jonathan entered the room and was
immediately caught by them. He immediately pointed to them, swinging his point over the
set of three. He did not point to anything else in the room.
Given the interactive nature of language acquisition, it is not surprising that mothers
very quickly come to exploit the new skill in pointing. The most outstanding of such is
by the incorporation of pointing and accompanying vocalization into rhetoric information
exchanges regulated by WH- interrogatives. Indeed, it is the ideal setting for the introduction of the forms of the locative. Typically, the mother will ask the child, Wheres
the? choosing as her referent some changing or otherwise attention-worthy object
like a light or aeroplane flying overhead. The child obliges by pointing, initially with an
indifferent vocalizationRichard used a variety of these, while Jonathan used either /Eh/
or
It is interesting to note that the child obliges by pointing, operating on some sort of
Gricean maxim that within limits, such requests should be met. But the style of meeting
them indicates that the exchange is understood as rhetorical. The point, in answer to such
persistent WH- questioning, becomes perfunctory and may omit altogether both vocalization and looking to the mother.
The child has his ways of exploiting his new skill as well, for he now occasionally incorporates pointing into well-established routines that he enjoys. It happens, for reasons that
will be apparent shortly, that we have been studying the development of give-and-take or
exchange formats. The reason is rather evident: language is itself an exchange routine and
we wish to examine how it is co-ordinated with other such routines. In any case, Richard,
shortly after the first appearance of pointing, inserted it into give-and-take games. But the
pointing is principally reserved for objects that involve a choice of alternatives or that are
ambiguous with respect to choice by being too distant for Richard to take directly. Two
examples are interesting. Richard had given his mother two cups. She placed one on her
head and the other in her lap, both out of reach. He pointed to the one on her lap and then
continued the point to the top of her head, vocalizing. We are not sure if he was demanding. His behaviour did not seem exigent. The other relates to the stabilized struggle over
our microphone for the audio-recorder (we use both video and audio, the former being
inadequate for phonological analysis). It is a forbidden object. Richard persisted in asking for it when it was lying on the rug within sight. He reached toward the microphone, too
distant to reach, pointed at it and vocalized, and looked at the experimenter. Let it be noted,
paradoxically enough, that this incorporation into action situations occurs early after the
appearance of pointing, is never frequent, and seems to decrease in proportion, as pointing
becomes established as a device for rather purer forms of information exchange.
We come now to the stage at which pointing and vocalizing begin to go through a
new transformation, concerned now with standard vocal labelling. Here we must begin by
picking up a point not developed earlierthe importance of distancing by the use of twodimensional representation of three-dimensional objects, animals, and persons. I refer of

course to the picture book, an enormously useful bit of technology in language acquisition.
When books are first introduced, the child typically treats them like other objectsto be
banged, carried, mouthed, clawed over. The first task is to get the child to look at the pictures without taking hold of the book as an objectthe attentional system and the manipulatory system seem initially to be in conflict. The first point is usually a forefinger dragged
across the page on which an object is depicted, usually terminated by a scratching movement. Then there appears a smiling, touching reaction to the pictureeither with the palm
or the fingers. The child then comes to sit back, looking, with excited vocalization, and with
recognition cries for familiar objects. Posture has changed from reaching and grabbing to
sitting back. The child now treats the book as a source of information and delight and will
fetch his book either spontaneously or in response to a WH- question from motherlike
Wheres Baby Bear?
Mother uses many attention-getting techniques initially to get the child focused. But
soon the child is able to direct attention himself and to point and vocalize at familiar pictures. Notably, the mothers own pointing behaviour becomes concentrated in book-reading sessions. At the same time, she is doing a great deal of labelling of indicated pictures.
There is, in the Richard corpus, a tenfold increase in book-picture labelling (in a half hour)
between 12 months and 15 months. These labellings are of highly familiar objects, most
usually, and there is an enormous amount of revisiting of the same pictures. What can be
readily discerned in the growth of book-reading is a distinctive cyclic pattern. The first
phase is characterized by the mother asking the child what a given depicted object is:
Whats that? or Wheres the piggy? or See Jack and Jill. The emphasis is not upon
labelling, but upon getting the child to point discretely to an object labelled by the mother.
If we look at Richard in this first phase (about 11 months), we find that his response rate
in reaction to different manoeuvres by the mother, where his response is defined as looking, a vocalization, smiling, or touching, is 75 per cent to her labelling, 50 per cent to her
Look, and none at all to her WH- questions. Any time the child vocalizes during this first
phase, she invariably accepts the vocalization as correct, with a Yes, thats right; thats a
rabbit. She is never corrective in tone or in vocabulary. During this phase, the child readily
lapses back into pounding or grabbing the book, etc. In the second phase, once the child
has achieved better control of inactive attention and when he is responding steadily to
the mothers labelling, the mothers WH- questions (previously not responded to at all) are
now the ones most likely to evoke response. Now, 79 per cent of mothers WH-questions
are reacted to, and labelling is only responded to 17 per cent of the time, with Look
yielding a 47 per cent response rate. His typical response to the WH- question is pointing
and vocalizing. There is a stunningly regular pattern that emerges. If the child points and
vocalizes, mother will virtually never ask a WH- question in response. She will either give
a confirmatory label or an elaboration. If, on the other hand, Richard points and does not
vocalize, the mother invariably asks a WH- question. What the child is being presented
with is an implied demand from the mother to complete a deictic gesture with a label. If
the two are not present, she will continue the dialogue until both have occurred. Note that
at this stage, the child vocalizes in strings of babbled syllables.
This brings us to the next phase in which, now, the mother starts imposing stricter
acceptance limits on the childs response in picture dialogue. She will, once she is sure that
the child will vocalize to each exchange, begin imposing the following constraint. Before,

if he vocalized, she would never use a WH- question in reply. Now, if the childs vocalization is non-standard or babbled-syllabic, she will ask a WH- and even correct his utterance,
and not give approval. She may lighten the situation by laughter, but the insistence is there.
She is now fishing for phonological hypotheses: the child is expected now to use her vocalization as a model. Of course, given that imitation is in the childs bag of tricks, whatever
imitation may be, the child obliges within limits.
In Richards case, by 15 months labelling is well established. It is interesting to note
that his labelling now begins to enter into other contexts in a rather interesting way. Again,
give-and-take routines provide an interesting tracer. At 17 months, Richard is very much
master of handing off objects, expecting them back, and possibly by now expecting objects
in exchange. These exchanges are often marked with vocalizations at the point of intention
and of accomplishment (reminiscent of DeLagunas proclamations of intent and accomplishment), but the vocalizations are babble-like and non-referential. Now, for the first
time, he hands an object to a partnerin this case one of the experimentersand labels the
object at the same time. We do not yet know whether it is important to note that the object
in this case was a flat plastic lid being used and exchanged as a pretend hat, the vocalization
being an ill-formed hat. There is also labelling for seeming self-consumption. In the same
session, Richard comes unexpectedly upon a biscuit lying on the floor. He stops, takes it to
his mouth, and vocalizes biki without looking at anybody. Again, it may be an instance of
marking the unexpected.
The history of labelling is itself rather intriguing, for what is striking about it is that
many nominals during the one-word phase appear and then disappear. Bloom (1973), for
example, notes that while her Allison had a cumulated or diary vocabulary at 16 months
of some fifty words, there were never more than twenty or twenty-five of them in play during any single week. The others had gone underground, sometimes to reappear in original
form, sometimes transformed, as dog disappearing and then reappearing months later as
bow-wow. It may well be that these labels are used initially for utterance as a means of
marking. Once the object named falls into its place, becomes part of the context, its label
drops out. Greenfield and Smith (1976) notice in a consonant fashion that words are used
for those aspects of a situation where there is still some uncertainty. Thus, an object being
pursued may be named. Once it is in hand, it is virtually never named. In some deep sense,
then, the procedures of language are taking over functions previously fulfilled by other procedures for drawing attention to what is new in contrast to what is presupposed or given.
But the step forward is a very genuine one for it permits a generalizability in use that was
simply not there before with such primitive devices as pointing, showing, or marking in
other ways.
We may return to Deweys point for a moment. Plainly, the task of language acquisition
involving labelling requires a form of interaction that stimulates the child to take on linguistic functions on his own. As Dewey says, communication by itself does not accomplish
anything. In so far as the dialogue between mother and infant succeeds in getting the child
to fill his role in exchangethrough labelling, through responding to WH- questions, etc.,
etc.the child is in fact learning not so much a language, as how to proceed in achieving
certain ends by the use of language. The input is not a corpus; the output is not a grammar.

We come finally to the pragmatic function, and unfortunately time is insufficient to say
much about its development from the point where I left it last spring. In my previous
paper, I commented upon the importance of the emergence of task structures, often carried
out in play, that had three crucial featuresdivision of labour, externality, and constraint.
Division of labour is clear enough, and involves role differentiation between two or more
partners. Externality implies simply the continued existence of an interaction format that
can be resumed, re-enacted, etc. It exists and is, indeed, labelable. Constraint refers to a
rule system or a system of informal demands for each to operate in a way expected by the
partner in the format. It was proposed earlier that these interaction formats provide the
child with working concepts relevant to later grammar like Agent, Action, Object, Recipient of Action, and the like. In a sense, they provide clues for cracking the grammatical procedural code of the language; and they may possibly also provide hints about order rules
which determine that initial grammars in all early language patterns studied involve the
orders: Agent-Action, Action-Object, Agent-Object, or Agent-Action-Object, and rarely
any reversals of these orders.
What we can report about later development of these well-practised patterns is that they
continue to provide segmentation points for the occurrence of the childs vocalizations:
vocalizations appear most often at the initiating position and at the completion pointas
I commented earlier, as intention and accomplishment markers. With time, they begin to
appear as well at crucial intervening points, as when the child, looking for an object to
hand to mother in an exchange, vocalizes at the point of finding the object, or when the
child develops an operator like up to characterize the action desired. These action formats
seem to provide the coherence in vocalization that early on led linguists like Bloch and
DeLaguna to urge that single words were holophrases, sen-tence words, or mot phrases.
In no sense can these formats be regarded as action grammars, for they do not specify
how sentences should be formed. But they provide a steady matrix into which indifferent
vocalizations can first be put, and then standard morphemes. Bloom (1973) brings to our
attention the fact that when successive one-word utterances appear, each separated by a
proper pause, they begin to take on the property of connectedness between successive
singlets: Mummychairread as the child goes about the task of getting a book down and
recruiting mother into sitting in a chair and reading a book with her.
We can perhaps add to this picture with a brief analysis of negation as it appears in the
well-practised routines that join mother and child. One such has to do with going after
objects with sensitivity as to parental permissionor, better, parental prohibitionas in
approaching a forbidden object and saying nonono with a characteristic prosody. There is
another format involving putting objects into a container, the objects often being handed
to him by his mother or asked for by Richard and handed by her. At 17 months, after
the establishment of early prohibitive negation, the form is generalized to the dimension
of appropriate-inappropriate. Objects that cannot be fitted into the narrow opening of his
favourite videotape box are now commented upon with head-shaking and no. An extraordinarily wide category of things that go, things that are permitted, things that fit is being
handled by negationall related to the action structure in which the child is operating.
I would only say one thing in conclusion. The process of acquiring language is very strongly
mediated by a push to manage various extra-linguistic functionsregulating joint atten-

tion, relating to others, getting certain tasks completed. The child manages these initially
by a diversity of communicative procedures. Mastery of each procedure seems to produce
a change, even an abandonment of the procedure. There is oftenas in Blooms recent
work and in McNeills discussionsa question as to why the child goes on to more subtle
procedures, such as those contained in grammatical discourse. Our evidence would indicate that, among other things, the mother raises the antes, makes her response somewhat
contingent upon the use of the more powerful methodsas in our examples of pointing
being replaced by labelling. But there are undoubtedly, as Bloom and McNeill argue, reasons of economy as well. Better techniques do lead to less ambiguous outcomes. But the
outcomes, I think, are not just linguistic, but extra-linguistic, relating to experience, task
prosecution, etc. I think Dewey had a point, and I am not saying this out of courtesy to the
great man we are celebrating in these lectures. Communication works because it evokes a
function to be fulfilled in the child. He may do a certain amount of rather blind imitating,
although that is not clear. Mostly what he is doing is entering into joint operations with
his mother-tutor and learning to use communication to bring those operations offwith
gentle pressure from the speech community. How much of this is innate? I dont know. I
suspect a certain amount of it islike using vocalization so readily, and entering so easily
into interaction, etc. But as another great pragmatist, William James, said, the innate occurs
only once; after that it becomes subject to the effects of experience.
Note
We are most grateful to Renira Huxley for her helpful comments.
References
Austin, J.L. (1962), How to Do Things with Words, Oxford University Press.
Bloch, O. (1921), Les premiers stades du language de lenfant, J. Psychol. Norm. Pathol., 18, pp.
693712.
Bloom, L. (1973), One Word at a Time: the Use of Single Word Utterances before Syntax, The Hague,
Mouton.
Bruner, J.S. (1975), Early language acquisition: a spiral curriculum, Doris Lee Lecture, University
College, London. For an extension of this paper see Entry into early language: a spiral curriculum, Charles Gittins Memorial Lecture, University College of Swansea, 13 March 1975.
Chomsky, N. (1957), Syntactic Structures, The Hague, Mouton.
Chomsky, N. (1962), Explanatory models in linguistics, in E.Nagel, P.Suppes and A.Tarski (eds),
Logic, Methodology and the Philosophy of Science, Stanford University Press.
DeLaguna, G. (1927), Speech: its Function and Development, New Haven, Connecticut, Yale University Press.
Dewey, J. (1926), Thinking in education, in Democracy and Education, New York, Macmillan.
Ferrier, L.J. (1975), Dependency and appropriateness in early language development, paper presented at the Third International Child Language Symposium, 35 September.
Greenfield, P. and Smith, J.H. (1976), Language beyond Syntax: the Development of Semantic Structure, New York, Academic Press.
Grice, H.P. (1975), Logic and conversation, William James Lectures, Harvard University 19678,
in P.Cole and J.Morgan (eds), Syntax and Semantics, vol. 3 (Speech Acts), London and New York,
Academic Press.

Halliday, M.A.K. (1975), Learning How to Mean: Explorations in the Development of Language,
London, Edward Arnold.
McNeill, D. (1974), Semiotic extension, paper presented at the Loyola Symposium on Cognition,
30 April, Chicago, Illinois.
3
Deweys theory of interest
Alan R.White
1
Though one usually associates Deweys views on interest with his educational theories,
they seem to have received essentially their full and final shape in his earliest and preeducational writings on psychology and moral philosophy. Moreover, they both closely
resemble those current among his immediate predecessors and contemporaries, such as
Herbart, Mill, Stumpf, Stout, James and Titchener (cf. Arnold, 1906) and in some respects
anticipate later psychologists, e.g. in their emphasis on interests as motives and on the personality or self as a system of interests (cf. Berlyne, 1949). Yet all these views of interests
deriveas I think Dewey saw quite clearlynot from any empirical work in psychology
or education, nor from any moral principles, but from philosophical analyses of the notion
of interest. What, therefore, I shall try to do is to see how far Deweys conclusions about
the nature of interest are justified. Others, more skilled than I in the appropriate field, have
elsewhere described how far his and other current educational views are supported by the
facts about interest (e.g. Wilson, 1971).
Dewey and his contemporaries sought a clue to the nature of interest in the three ideas
of attention, feeling and object. Subscribing in his very first book on psychology (1886, pp.
1517)1 to the then fashionable triad of the cognitive, the affective and the conative aspects
of mind, Dewey regarded consciousnessthe presupposed topic of psychologyas
involving a knowledge-obtaining activity accompanied by a self-affecting feeling. This
activity is attention and this feeling is interest. Later in the same book (pp. 2768) interest
is said to be the connection of both these elementsi.e. the activity of attention and the
self-affecting feelingwith the object to which the attention is directed and which arouses
the feeling. The combination of these three elements in interest was reiterated five years
later in his first book on ethics (1891, p. 305). There, however, he also committed himself
to what I shall later argue is an unnecessary piece of metaphysical nonsense. Interest, he
said, is the union in feeling, through action, of self and an object. An interest in life is had
when a man can practically identify himself with some object. Five years later still, his
first venture into educational theory (1896, p. 90) takes the form of an application of his
psychological and moral analyses of interest. Here reappear the analysis of interest into
three elements and the identification of the person who is interested and that in which he is
interested; the latter bolstered up by an appeal to two logico-linguistic features of the word
expressing the notion of interest, whose force I shall discuss later.
I believe that Dewey was basically rightas were also his contemporaries and immediate predecessorsin his analysis of interest in terms of a subject interested, an object in

which he is interested, a relation of attention between the subject and object and a feeling
which the subject has when he gives this attention to this object. Where, as I shall now
argue, he, like his contemporaries, goes wrong is at various times, perhaps from excess of
zeal, to succumb to the temptation of identifying interest with one or other of its elements
to the exclusion of the others. The identification of interest with attention and its identification with feeling are fallacies which Dewey shared with most of his contemporaries. His
own peccadilloes consisted in the identification of interest either with the subject of interest, that is, that which has the interest, or with the object of interest, that is, that in which
the interest is had, or with something which he pictured as the union of both the subject
and the object.
To see this, let us turn to an examination of the notion of interest and its elements.
2
Dewey waivered (cf. 1912, p. 472) between saying, in some places (e.g. 1896, p. 76), that
interest is the same as attentionan identification which he sometimes (e.g. 1891, p. 305)
attributed, quite wrongly, to ordinary usethat it is a necessary condition of attention,
so that every instance of attention is due to interest, and that it is a sufficient condition of
attention, so that every instance of interest is manifested in attention. At exactly the same
time, Stout was arguing that we cannot be feeling an interest in an object without attending to it and we cannot be attending to it without feeling some interest in it (Stout, 1896,
pp. 2245), while a little later McDougall concluded that Interest is latent attention, and
attention is interest in action (McDougall, 1923, p. 277).
Clearly this will not do.
First, the notion of interest can be used both dispositionally and episodically, whereas
the notion of attention is used only episodically. Interest is something we can either have or
take (or show) and to be interested signifies either having or taking (or showing) an interest. To be too interested in the faults of men to see their virtues is to be of a certain disposition, while to be too interested in ones book to hear the doorbell is to undergo a specific
experience. While we cannot take (or show) an interest which we do not have, we can have
an interest without at a given moment taking (or showing) it. There is no contradiction in
saying that I have been for years interested in limericks, though for months I have taken
(or shown) no interest in them nor given any of my attention to them. I can be interested in
thousands of things, but take (or show) an interest in only a few at a time. Someone may
observe me taking (or showing) an interest, but not observe me being interested, in what
lies around me. Ceasing to take (or show) an interest is not necessarily losing interest.
Furthermore, one can be dispositionally interested in doing something one is not at
the moment doing, but not taking (or showing) an interest in doing what one is not at the
moment doing. Only when one is taking (or showing) an interest, therefore, can the object
of ones interest necessarily be affecting one.
Now Dewey clearly recognized at different times the dispositional use of the notion of
interest, that is, the idea of having an interest which one is not at the moment manifesting,
and the episodic use, that is, the idea of taking (or showing) an interest; for, on the one hand,
it was a central tenet of his theory that a child is possessed of, indeed, almost constituted
of, a set of interests which it is the job of a good teacher to arouse in order to get the child
Deweys theory of interest 25

to take an interest in or to pay attention to the required features of the lesson (e.g. 1896
passim). Indeed, I shall argue that Dewey so distorted this dispositional use of the notion
of interest as to confuse having various interests with being of a certain character of mind.
On the other hand, he was always so impressed by the fact that interest is something we
can take as to suppose (e.g. 1891, p. 305; 1896, pp. 90, 95; 1913, p. 16) that interests are
necessarily active and result in activities, all of which involve paying attention. Because
interest is both dispositional and episodic these two views are perfectly compatible, and not
unnaturally Dewey was able in practice to emphasize either one or the other. It was only
his apparent failure to see that there were two different uses here that led him to assert, on
some occasions, that interest is necessarily an activity and on others that it is necessarily a
constituent of ones mind like a desire, instinct or feeling.
More important, however, is the fact that attention is differently related to dispositional
interests and to episodic interests because attention itself is only episodic. It is something
we can only pay or give and not something we can possess without displaying. Attention,
unlike interest, cannot be latent and waiting to be aroused. Even the attentive man is necessarily paying attention. Clearly, therefore, to be (dispositionally) interested in something no
more implies to be paying attention to it than it implies to be actually taking (or showing)
that interest which one has in it. A fortiori, interest cannot be identical with attention.
Nor will an identification of interest and attention work even for taking (or showing)
interest. Though one who is at this moment taking (or showing) an interest in something
must be paying attention to it, he can be paying attention to it without taking (or showing)
any interest in it, and, a fortiori, without having any interest in it. It is both logically different and socially more flattering to say that you read a colleagues book with deep interest
than to say that you read it with close attention. Neither having nor taking (or showing) an
interest is something we can decide, resolve, refuse or reasonably order someone to do as
we can decide, resolve, refuse or reasonably order someone to pay attention. Being here
and now interested or uninterested is something we cannot help. To keep your attention on
something, e.g. to keep staring at the man opposite, is to continue doing what you have
been doing, but to continue to take an interest in it is to continue this in a certain frame of
mind. It is unfair to blame someone for not taking (or showing) any interest in a way that it
is not unfair to blame him for lack of attention. Certainly, one may try to take (or show) an
interest in something, but this is rather like trying to feel sorry for someone: however hard
we try, we may not succeed. This is why we often say to someone that he might at least look
interested or pretend to be interested, just as we say that he might look or pretend to feel
sorry. People who want to pretend to have an interest which they do not have or to pretend
to be taking an interest which they are not taking, pay attention to certain things and try to
look interested, just as people who want to pretend to an affection they do not feel ape the
behaviour of the affectionate. Contrariwise, to conceal either your interest or the fact that
you are taking an interest in something is not to keep any activities hidden in the way that
concealing the fact that you are attending is to conceal an activity.
When our attention wanders from our task, we become absent-minded or we forget it;
when our interest wanes, we have grown tired of it. Both the man who loses interest and
the man who ceases to take (or show) any interest may, but do not necessarily, stop paying
attention, though they will no longer pay attention in the same frame of mind. Attention
is glued or fastened on something, interest is sustained or retained in it. There can be rea-

sons for paying attention to something other than those given in terms of interest in it or
in something else. People can be made to pay attention by threats and warnings or the call
of duty. Most commonly, however, it is some feature of the object of our attention which
makes us attend to it; what draws my attention to the clock is its suddenly ceasing to tick
or its going off with a bang (contrast 1886, p. 127). But whereas both having an interest in
something and taking (or showing) an interest in it can, though they need not, explain paying attention to it, paying attention to it cannot explain either having or taking an interest
in it. We may attend because we are interested; we are not interested because we attend.
To say that we are interested because we are inclined to attend is not to explain our interest
but, as we shall see, to explain the meaning of the word interest.
Whenever Dewey had in mind a childs being dispositionally interested in something,
he realized clearly that it need not be at that moment paying attention. Indeed, he stressed
that it was just such an interest that the teacher had to concentrate on to capture and hold
the childs attention. Equally, he realized, that a childs attention could be brought to bear
on something without first arousing its interest, either because its attention was caught by
one of those features, such as prominent size or loudness, which instinctively and naturally
draw human attention (though he often seems to assume that even these are necessarily
interesting, e.g. 1886, pp. 121f.; contrast 1896, pp. 12930) or because its attention was
compelled by threats of punishment or enticed by promises of reward (e.g. 1900, pp. 923;
1896, pp. 839; 1913, pp. 334). He even allowed that a childs attention could at any
moment be divided between those things that interested it and those that did not, the latter
being, he thought, only external mechanical attention (1896, pp. 857; 1900, p. 93; 1913,
pp. 911; 1916, pp. 2078). Indeed, for Dewey the greatest and most interesting problem
in the analysis of interest was the alleged choice between getting attention by arousing
interest and getting it by other methods (e.g. 1896, pp. 76f.). The fact that he thought, as
we shall see, that the necessity for this choice rested on a misunderstanding of the relation
between the subject and the object of interest does not show that he thought that it was
impossible to have attention without interest.
Nevertheless, despite these contexts in which his whole problem and argument presupposed the possibility of paying attention to something without taking (or showing) any
interest in it, he seems generally to have assumed and frequently committed himself to the
mistaken declaration that, for example, attention is always realising some interest of the
self (1886, p. 138; cf. 22, 131) and that It is psychologically impossible to call forth any
activity without some interest (1913, p. 2; cf. 1886, pp. 1819; 1916, p. 408). Only attention which was caught and never attention which was given was allowed to be interest free
(e.g. 1886, p. 132). The notion of interest was even built into his definition of attention as
its basis (18,86, p. 133).
If, then, attention is so far from being identical with interest that to be interested in
something neither implies nor is implied by paying attention to it nor is taking (or showing) an interest in it implied by, though it does imply, paying attention to it, how, if at all,
is interest related to attention? And did Dewey see this?
Several times he refers to a notion which, I think, gives us the desired clue. In 1886 (pp.
1213), though he does not always distinguish between catching our attention and attracting it, he equates those features which make an object interesting with those features
which attract the mind and the interest of an object with its attractive power, while in

1913 the lack of interest of a programme of studies is identified with its lack of power
to hold attention.
As we have seen, attention can be demanded but interest has to be aroused. Conscientious, deliberate, willing and unwilling, half-hearted and reluctant are appropriate
to attention, but not to interest, for the idea of interest contains the idea of attraction. The
bored spectator has difficulty in keeping his eyes and his mind on the game, whereas when
he is interested his difficulty is to take and keep them off it. The man who loses interest
does not necessarily stop paying attention; but he does lose any inclination to pay it. Interest in something is an inclination to give ones attention to it often coupled with a disinclination to pay attention to anything else. When interest passes over into fascination, the
inclination to pay attention becomes an inability not to pay it. Fascinating is the superlative of attractive. To be fascinated by a woman is to be unable to take your eyes or your
mind off her. The club bore, on the other hand, is one who, by his insistence, compels us
to attend against our inclination.
3
The second element which occurs in Deweys and his contemporaries analysis of interest
is feeling (cf. Arnold, 1906). By feeling Dewey meant the way in which any conscious
creature is affected by that of which it is conscious; so that, in this wide sense of feeling,
there is an element of feeling in every mental phenomenon, whether it is, for example,
belief, knowledge, perception or thought on the one hand, or desire, emotion, sensation or
mood on the other. As Dewey put it, consciousness and experience are never colourlessly
intellectual (1886, pp. 16, 276, 286). In his very first work (1886) Dewey called this feeling or emotional side of consciousness, this way in which the self is affected, interest.
Hence, he thought not of interest as a feeling in the sense in which we might say that fear
and pity are feelings whereas belief and perception are not, but of the word interest as a
synonym of the word feeling. So the third of this book devoted to feelings usually speaks
indifferently of feelings and interests. Later, in the same book (1886, pp. 2767), however, and in his subsequent writings, he distinguished between this mental affection as
mere or bare feeling, and reserved interest for the combination of this feeling with an
object and an attentive activity. His early identification of interest and feeling, however,
persisted so far as to lead him throughout his life to think of most object-directed feelings
as interests. He continued to subscribe to his early view that every object that comes
within our experience gets some emotional colouring, as it helps or hinders that experience. It thus gains a special and unique interest of its own (1886, p. 286). So he speaks
(1886, pp. 277f.) indifferently of personal, intellectual or aesthetic feelings and personal,
intellectual or aesthetic interests (1886, chs XIVXVI). He equates (1886, p. 340; 1891,
p. 305) love with interest in a person and characterizes fear and dislike as the obverse
aspect of interest (1900, pp. 445). He also assimilates interest and desire, either holding
that interest is the same state of mind as desire, but with possession of its object (1891,
p.305; 1886, p.361), or holding that desire is properly mediated interest (1896, p. 109).
For him, being desirable is being interesting (1886, ch. XVIII), a conflict of interests is a
conflict of desires (1916, pp. 2089) and the relation of interest to effort is that of desire

to effort (1912, p. 473). To relate ones teaching material to the interests of the child is, for
Dewey, to relate it to his urgencies and needs (1900, p. 23).
Now, whether or not Dewey is correct in supposing that everything of which one is in
any way conscious necessarily affects one and, therefore, gives rise to what he calls a feeling, it is quite clear that, in the first place, being conscious of something no more implies
taking an interest in it than, as we saw, does paying attention to it imply taking an interest
in it. Second, and more relevantly for our purpose, though interest is certainly something
we can, though not necessarily something we must always, feel, it is not in every sense of
feeling that we feel interested in things.
Feeling interested is obviously not a perceptual feeling like feeling a hole in my
pocket, or an exploratory feeling like feeling for a light switch. Neither is someone who
is interested in what he is doing necessarily having any sensations, faint or acute, steady
or intermittent, localizable or general; such sensations would distract him from the object
of his interest. An itching in his fingers might be a sign of interest, it is not the feeling of
interest itself.
Moods such as happiness, depression, frivolity, sentimentality, are commonly called
feelings. But interest is not a mood, for, unlike a mood, it has, as Dewey stressed, a definite
object; nor does it colour ones actions and feelings as a mood does. An apparent objection
to this denial that feeling interested is feeling or being in a certain mood springs from the
fact that we can certainly call feeling, or being, bored being in a certain mood and also say
that feeling bored by something is the contrary, though not the contradictory, of feeling
interested in it. The answer to this objection is that bored is used to cover both the mood
which consists in a general ability to feel interested in anything and the specific inability to
feel interested in a given topic.
Nor is feeling interested an emotion or stirred-up state like feeling excited or thrilled,
agitated or surprised. You cannot be beside yourself or speechless with interest; nor does
increasing interest disturb your concentration as mounting excitement or anxiety may.
What one feels when one feels interested in anything is, I think, an inclination to attend
to it. It is what Baldwin long ago called an impulse to attend and Stumpf a Lust am
Bemerken. To feel interested in something to which, for whatever reason, one does not
or cannot pay attention is to feel an inclination to which one does not or cannot yield. The
difference between the man who feels an interest in what he is giving his attention to and
the man who does not is that the former is, while the latter is not, giving an attention which
he feels inclined to give.
4
For Dewey the ubiquity of interest is due to its invariable presence not only and not mainly
as the resultant way in which, that is, the feeling with which, one is affected whenever one
is conscious of anything, but also as a motive for every action. Immediately after declaring that every object that comes within our experience gets some emotional colouringa
special and unique interest, he adds an object, as soon as it has become interesting,
becomes an end of action in itself (1886, p. 286; cf. pp. 317, 340). The role of interest as
a motive is for Dewey much more important than its role as the feeling that accompanies
consciousness. He emphasized (1896, p. 120) this as the main difference between his psy-

chology and that of Herbart whose views on interest he otherwise finds very appealing. It
is, moreover, the leading role which has been assigned to interest in more recent psychology (cf. Berlyne, 1949). Every impulse and habit, says Dewey, that generates a purpose
having sufficient force to move a person to strive for its realisation, becomes an interest
(1913, p. 90; 1896, p. 112; 1916, pp. 1523). Indeed, just as he tended to move from the
view that interest is a feeling to the view that the word interest is a synonym for the word
feeling, so he tends to move from the view that every motive is an interest to the view
that the word interest is a synonym for the word motive. Thus, motive is sometimes
(e.g. 1908, pp. 3212; 1896, p. 112) defined as those interests which form the core of the
self and supply the principles by which conduct is to be understood; motive and interest are used interchangeably (1908, p. 322) and the motive force of an end is declared
(1913, p. 60) to be an equivalent expression to the interest that the end possesses. Lists
of motives are called lists of interests (1913, p. 62), his doctrine of the identity of self and
object is said to be the key to understanding the nature of both motives and interests (1908,
p. 319), both motive and interest are said (1908, p. 322) to be applicable either to the
desire of an object or to the object desired, e.g. the desire for money and the money itself,
and his dictionary article on Interest opens with the declaration that: The doctrine of
interest in education is a sort of shorthand expression for a number of different motives
(1912, p. 472).
Certainly, an interest in something like stamp-collecting provides a possible motive
for action; but Deweys equation of interests with motives cannot be right, both because
someone can have a motive for doing something which is not an interest in it and can have
an interest in doing something which is not his motive for doing it. Thus, to categorize
jealousy, greed, revenge, love, hatred, ambition or patriotism merely as interests is to take
far too anaemic a view of human passions. On the other hand, someones interest in, for
example, stamp-collecting or philosophy need not on a particular occasion be his motive
for, for example, buying stamps or reading a philosophy book; perhaps he was buying the
stamps or reading the book for a friend. Dewey himself on occasion allowed that motives
could be supplied by such things as needs, cravings and demands (e.g. 1900, p. 40). Even
less correct is Deweys suggestion that the terms interest and motive are synonymous.
Interests, but not motives, can be aroused, excited or awakened; they can subside, vanish,
or be lost; they can devour and consume. Motives, but not interests, can be plausible or
far-fetched, ulterior or involved. An interest is one of the things that can be a motive, just
as a car can be a gift or a house an asset, but interests and motives are no more of the same
category of things than cars and gifts or houses and assets.
What I think may have misled Dewey into his identification of interests and motives
or even into his view that interests always motivate are, first, his equation of interest and
desire, second, his assimilation of the dispositional use and the episodic use of interest
and, third, his failure to distinguish doing something from (or for) interestwhich was
his main topicfrom doing something with interest.
First, though Dewey usually directly equated motives with interests, he also equated
motives with desires (e.g. 1886, pp. 3668; 1900, p. 79) and, as we saw, desires with interests, and, therefore, indirectly motives with interests. But the equation of interest and desire
is no more correct than that of interest and motive. Though someone who feels interested
in something may not merely feel inclined to give it his attention but also feel a desire to

do so, it does not follow, and is not true, that an interest is a desire for anything more than
to give ones attention to it. It would be too strong to say that a story which had aroused
my interest had aroused my desire and too weak to say that a desire for survival is just an
interest in it.
Turning, second, to the way in which an interest undoubtedly can explain what one
does, we must remember the distinction between (dispositionally) having an interest and
(episodically) taking (or showing) an interest.
Taking (or showing) an interest in, and hence paying attention to, something on a particular occasion may be due to and explicable by the interest that one has long had in it
or in this sort of thing. Having an interest, as we saw, is being disposed to take (or show)
an interest and, hence, to pay attention. In such an example we are explaining a particular
manifestation of a disposition by the disposition itself; we are saying that someone did X
on this occasion because he is disposed to do X.
Interest, in this dispositional use, can provide a reason for engaging in any activity and,
therefore, for paying attention. Indeed, the sorts of activities which can be said to be done
from interest must be such as involve paying attention. Like desire, the sorts of reason
that interest thus provides for doing something are twofold. You may do X or X-like things
because you are interested in X or because you are interested in Y. In the former case
you would say you had done X just from interest. For example, an interest in archaeology may be the reason why someone pays attention to archaeological matters by reading
books, attending lectures and joining in conversations on the subject. The sort of reason
that interest provides for paying attention to something or for engaging in any activity is
similar to that which jealousy provides for certain kinds of behaviour or that brittleness
provides for breakages. When you wish to attract someones attention you try to provide
the sort of thing which is liable to do this, something in which he is interested, just as when
you want a big smash you get something liable to break, something brittle. Advertisers
hoardings show pictures of pretty girls because, as a matter of definition, sex-interested
males will look at them. We are familiar with the anecdotes of psycho logy textbooks (cf.
Dewey, 1886, p. 135) whose point is that people of different interests, when placed in the
same situation, pay attention to and notice different features. On their communal walk the
geologist notices the rock formation, the botanist the plant life, the painter the landscape,
while the philosopher fails to notice anything. But the point illustrated is not, as psychologists sometimes suppose, a factual discovery about human behaviour, not something for
which we need experimental evidence; it is a logical truth about the relation of the notions
of attention and notice to that of interest.
We can be interested not only in objects and activities, but also in becoming or being or
attaining or bringing about something. Ends as well as deeds can attract us. Thus an interest in X can provide a reason for doing Y. The student who is interested in improving his
French has a reason for visiting France, buying French newspapers, seeing French films.
We often attend to something for its own sake, because we are interested in it, but our
reason for attending can also be that we are interested in finding out or discovering so and
so. Curiosity is this sort of interest. In these cases of doing X from interest in Y we do not
describe ourselves simply as doing X out of (from, for) interest.
Although Dewey did not go so far as to make the mistake of saying that when we do
X out of interest in Y, for example because X is seen as a means to Y, we, therefore, do

X itself out of interest, he does seem sometimes to have supposed (e.g. 1896, pp. 1001;
1913, pp. 257) that X itself necessarily becomes interesting to us; indeed, that the interest
which originally attached only to the end now attaches also to the means. It is in this way,
he thought (e.g. 1896, p. 97; 1900, pp. 436, 923; 1912, p. 474; 1916, p. 149), that the
successful teacher makes things interesting for the pupil and not by any merely external
sugar-coating of the pill. To make it interesting by leading one to realise the connection that
exists is simply good sense; to make it interesting by extraneous and artificial inducements
deserves all the bad names which have been applied to the doctrine of interest in education
(1916, p. 150). So convinced was Dewey that whatever was seen as a means to something
in which one was interestedor indeed something which was in any way desiredwas,
therefore, itself interesting that he several times even used the etymology of interest from
interesse (to be between) to prove his point. To be between the agent and his end,
he said (1916, pp. 14950; 1896, p. 99), and to be of interest, are different names for the
same thing.
Certainly, things, previously indifferent, can and do become interesting when their connection with something in itself interesting is seen (e.g. 1896, p. 100; 1900, pp. 847; 1912,
pp. 4734; 1913, pp. 22, 3841). But this is not necessary. To take Deweys own example
(e.g. 1896, p. 100), surely the fact that a man with a family to support has a new motive
for his daily work does not entail that he finds that work any less of a drudgery. Students
whose interest in foreign literature leads them to learn the irregular verbs of its language
do not necessarily find this learning interesting. An interest in X can make us pay attention
to Y without making us take an interest in Y.Dewey, I suspect, was here guilty of the logical error of supposing that if one is interested in, for example, providing for ones family,
and providing for ones family takes the form, on some occasion, of disagreeable drudgery,
therefore one must, on that occasion, be interested in disagreeable drudgery.
When what we do, including the paying of attention to something, is explained not by a
continuing interest we have but by an interest we are here and now taking (or showing) or
feeling, our explanation is of a different kind. For, since to feel interested is to feel inclined
to pay attention, then to pay attention to something because we feel interested in it is to do
what we feel inclined to do; our attention satisfies our desire to pay attention. It is when
the play is dull that, as Aristotle remarked (Nicomachean Ethics, X 5), we turn to our box
of chocolates.
It follows from this relation of interest and attention that the statement He was interested in so and so is what Ryle has called a mongrel-categorical relative to the statement He paid attention to so and so. A reporter might say of a distinguished visitor at
an exhibition either that He paid particular attention to the toy stand or that He took (or
showed) a particular interest in the toy stand. The description in terms of interest gives an
explanatory-cum-predictive account of the same event as the description in terms of attention. It can, therefore, replace it as a description, serve as an answer to the question why he
paid attention, and allow us conditionally to predict a continued attention to the toy stand
or things related to toys. Further, one can look or listen, examine or study, read or discuss,
with attention or with interest.
The third oversight which may have led to Deweys identification of interest and motive
is that between doing something from (for, out of) interest and doing something with
interest. Whereas from interest gives the reason for someones doing something, With

interest states the accompanying result of his doing it. Hence purely, solely, merely
which are used to emphasize the uniqueness of a reasoncan accompany from interest
but not with interest, while little, much, sharp, increasingwhich characterize the
degree of anythingmay accompany with interest but not from interest. Similarly, a
non-continuous action, such as asking a question, coming to a meeting, going to Greece,
can only be done from and not with interest; whereas a result, like seeing or hearing,
coming across or learning something, may be with but not from interest.
A person who reads a book from or for interest shows his interest by doing what
he does. His interest explains his reading of the book and allows us to make conditional
predictions about the doings of other things of the same sort. No such explanatory-cumpredicative account is given of the reading of a man who is described as reading with
interest. Nor does because he is interested in what he is doing explain why someone
does what he is doing, however well it may explain why he does not do something other
than what he is doing or why he continues to do what he is doing or why he pays attention to what he is doing. My interest in the philosophy book I am reading may explain
my not giving it up or my not answering the telephone or my attention to my reading. My
interest in the book does not, however, explain my reading the book, though my interest
in philosophy may.
5
Deweys theory of interest was, for the most part, that common in the philosophy and psychology of his day. There is, however, one major detail which seems to have been peculiarly
his own and which I have with deliberate abusiveness called metaphysical nonsense. It is
his identification of the person interested with that in which he is interested or, as he sometimes puts it, the union of self and object (e.g. 1896, pp. 83f.; 1908, p. 321; 1912, p. 472;
1913, pp. 14, 90, 95; 1916, pp. 1612). He appealed (e.g. 1896, pp. 83f.; 1913, pp.67)
to this identity to suggest that both those who advocated discipline, effort and threats and
those who advocated blandishments, bribes, and allurements to attract a pupils attention
make a common mis-assumption that the pupil and what he is interested in are two separate
things which need to be brought together by some means or other. A similar common misassumption with a similar solution underlies, he thought, both the ethical conflict between
the utilitarian emphasis on consequences and the opposing emphasis on the moral agent
(e.g. 1908, pp. 31519) and the traditional view that if someone always acts from some
interest, he therefore, always acts from self-interest (1908, p. 328).
But what on earth can Dewey have meant by saying that a personor, as he more usually says, a persons self, or mindwho is interested in something is either identical
with or forms a union with that in which he is interested? We can, I think, find at least four
logico-linguistic features which help to explain what he meant by this and why he believed
it to be so.
First, he regarded interest as a feeling which was integrated with the object that
aroused it, partly, I suspect, for the sort of reason that led Aristotle to identify pleasure
in an activity with the completion of the activity in which one takes the pleasure (cf.
1896, pp. 879; 1913, p. 12) and partly because he was struck by the fact that we use the
word interest, as we use pleasure, need, want, etc., both for someones interest,

pleasure, need, want, etc., and for the object of his interest, pleasure, need, want, etc.
Thus, philosophy and stamp-collecting are the interests of the man who is interested
in philosophy and stamp-collecting as the sights, sounds and tastes from which we get
pleasure are called pleasant (cf. 1886, pp. 276, 289; 1891, p. 305; 1896, p. 91; 1908,
pp. 3213; 1912, p. 475; 1913, p. 16; 1916, p. 148). Similarly, the value of an object
is, Dewey considered, our valuation of it (1886, pp. 276, 289; 1896, pp. 945). Beauty
is both in the eye of the beholder and in the object beholden (1886, p. 322). In short,
what is interesting, like what is pleasant, attractive, amusing, desirable, or what is a
need or want, is what people find interesting, pleasant, attractive, amusing, desirable,
or what they need or want. Now, emphasis on the fact that what one has an interest in is
called ones interest could tempt one to single out the object of interest from the three
elements of interestattention, feeling and objectand plausiblythough, of course,
quite incorrectlyidentify interest with that element. And Dewey does sometimes succumb to this temptation. But clearly none of this would go any way to show that either
the interested person or the interest he has is joined with, much less that it is identical
with, the object of his interest any more than the analogous feature of pleasure, need,
want, etc., shows an analogous union. It would be absurd to say that I am identical with
Plato, his dialogues, or his Theory of Forms just because I have an interest in all of these.
And it would be equally absurd to conclude from this that since I am identical with each
of my interests, then all of my interests are identical with each other.
Second, Dewey appealed to the etymological derivation of interest from the Latin
interesse, meaning to be between, to argue that interest was a bond which annihilated
the distance between the interested person and the object of his interest (1896, p. 91; 1912,
p. 472; 1913, p. 17; 1916, p. 149). But even if interest did join the two, it would not make
them identical.
Third, Dewey seems to have taken too literally various metaphorical expressions of
the way in which our interests occupy our mind. Thus, the interested person is said to be
engaged or occupied, taken up with, concerned in, absorbed by, carried away by
his material (e.g. 1896, pp. 83f., 91; 1913, pp. 17, 65, 90; 1916, p. 148); he finds himself
at home in it (1912, p. 472). From this, Dewey concludes that what he calls completeness of interest is whole-hearted identification with what one is doing (1913, p. 80). But
clearly to take these metaphors of absorbed attention literally as identification of someone
with what he is interested in is as if a philosopher were to argue that because a colleague on
Senate was, as we say, at one with him about a certain measure, therefore he was identical
with this colleague.
Deweys final type of argument for the identification of the interested person and the
object of his interest also stems partly from taking metaphors literally. On several occasions an explicit reason for the identification of the self with some object or idea is the
necessity of that object or idea for the maintenance of self expression (1896, p. 89; 1913,
p. 14). This is rather like identifying oneself with ones needs on the strength of the figurative phrase I need it to be myself. Related to this is the argument that because a man
who is interested in his work is said to find himself in the work, whereas a man with a
different set of interests would be a different kind of man, a man is, therefore, his interests
(1916, p.408, cf. p. 162; 1913, p. 43). His interests form his core (1908, p. 321); for interest defines the self (1908, p. 327; 1916, p. 408). Dewey held that, if we are psychically

awake at all, we are always interested in one direction rather than another (1896, p. 93;
1908, p. 320; cf. 1910, pp. 36f. on curiosity). And he, rightly, argued (1908, p. 322) that,
for example, hunger, benevolence, cruelty, and kindness, are not the names of entities which make us do things but of tendencies we have to do them. But, clearly, to identify
either someone or, more plausibly, his personality, character or mind, with his interest, as
we might reasonably, in Aristotelian or Rylean fashion, identify his mind with a set of intellectual and moral abilities, capacities, dispositions and functions, somewhat as we identify
a house with the bricks which compose it (1908, p. 316), is to identify his personality with
the interests which he has in things and not with the things in which he has these interests,
with his inclination to pay attention and not with that to which he pays attention. Here
Deweys argument points rather to the (mistaken) identification of interest with the subject
of interest than to the (mistaken) identification of the subject and the object of interest.
Only a failure to remember the distinction between my interest in birds eggs and the birds
eggs which are my interest could lead to the preposterous conclusion that I am somehow to
be identified with a collection of birds eggs.
It is the philosophical absurdity of these four logicolinguistic arguments for Deweys
oft-repeated identification of the subject and the object of interest that makes me dub the
identification a piece of metaphysical nonsense and makes me reject it as an attempt to
solve the disputewhich Dewey over-exaggeratedbetween those who look for an incentive to attention in discipline and those who look for it in interest.
6
Finally, just as Dewey was deflected by his digression into metaphysics from any serious
attempt to tackle this dispute, so his failure to examine deeply enough the relation between
ones interest in something and somethings being in ones interest deflected him from
any serious attempt to face the possibility of a conflict between them as rival criteria of
the purpose of education. He seems to have assimilated (e.g. 1916, pp. 148, 401) the idea
of some things being in someones interests to the idea of someones being interested in
that thing. Hence, to act in ones own interests or in the interests of others was to act with
an interest in ones self or with an interest in others and the possibility of conflict between
self-interest and the interest of others was treated as a conflict, not between what is in ones
own interests and what is in anothers interests, but between ones being interested in ones
self and ones being interested in others (e.g. 1891, pp. 3067; 1913, p. 88; 1916, pp. 148,
408). This assimilation was, perhaps, largely due to Deweys assumption that to be interested in someone entailed being interested in securing what was in his interests (e.g. 1891,
pp. 3067; 1908, p. 326).
But what is in a childs interests may not be what the child is interested in. Indeed, he
may have no idea whether something is in his interests or not, but he is hardly likely to be
so ignorant of what he is interested in. Things can be done, as Dewey himself writes, in the
interests of efficiency, fair play or justice (1908, pp. 330, 344), but such inanimate things
cannot be interested in anything. Hence, even if we make the childs interests the corner
stone of our educational philosophy, we are still left with the question: Should we do what
the child is interested in or what is in his interests?
Note
1 Dates given for Dewey are those of first publication; page references are to editions cited in the
References.
References
Arnold, F. (1906), The psychology of interest, Psychological Review, vol. XIII, pp. 22138, 291
315.
Berlyne, D.E. (1949), Interest as a psychological concept, British Journal of Psychology, vol.
XXXIX, pp. 18494.
Dewey, J. (1886), Psychology; pagination of 3rd edn, New York, 1894.
Dewey, J. (1891), Outline of a Critical Theory of Ethics, pp. 30411; pagination of The Early Works
of John Dewey 18821898, London, 1969.
Dewey, J. (1896), Interest as related to the training of the will; pagination of Educational Essays,
ed. J.J.Findlay, London, 1910.
Dewey, J. (1900), The Elementary School Record, Nos. 4, 5, 6; pagination of The School and the
Child, ed. J.J.Findlay, London, 1907.
Dewey, J. (1908), Ethics, ch. XV; pagination of 1932 edn, New York, 1959.
Dewey, J. (1910), How We Think, chs III (section 1), XIVXV; pagination of 2nd edn, Boston,
1933.
Dewey, J. (1912), Interest, in A Cyclopaedia of Education, ed. P. Monroe, vol. III, pp.4725.
Dewey, J. (1913), Interest and Effort in Education, Boston.
Dewey, J. (1916), Democracy and Education, chs X, XXVI (section 2), New York, 1960.
McDougall, W. (1923), An Outline of Psychology, London.
Stout, G.F. (1896), Analytical Psychology (2 vols), vol. 1, London.
Wilson, P.S. (1971), Interest and Discipline in Education, London.
4
The self in action
Martin Hollis
Students of human nature used to scour the wide world for their data. In these more
advanced days, however, they may prefer to consult the International Encyclopedia of the
Social Sciences. If so, they will be disappointed. There is no entry for Human Nature. But
then the unity of the International Encyclopedia is that of a shopping centre which just
offers a roof to cognate specialists. In some ways it is a poor replacement for the general
store, supplying articles which everyone needed whatever his special concerns. Those who
take the hint and try the older Encyclopedia of the Social Sciences will indeed find an entry
for Human Nature. It is by John Dewey and is as clear, concise and wide-ranging as one
could wish. Yet it contains a remark at odds with much of Deweys own work. It cannot
be doubted, he writes, that there are some limits to the modifiability of human nature and
to institutional change, but these have to be arrived at by experimental observation. His
point seems almost banal. It is that men are not wholly plastic and that social processes
neither shape nor explain them in toto. Yet it springs a twofold surprise, for, in the first
place, Dewey must have devoted a million words in a long life to casting doubt on the fixity of human nature and, in the second, was always too much of a philosopher to leave the
truth to experimental observation. I shall invoke his aid in seeing both what the truth about
human mutability may be and what sort of truth it is. The question concerns the role of the
self in the philosophy of human action and Deweys answer will show us the strength and
the snags of a powerful pragmatism.
My reason for turning to Dewey is not just that he is a noble, systematic thinker, full of
perennial insights, wise, shrewd and blessed with a power of vivid prose.1 Nor is it directly
the importance of his instrumentalism nor the influence of his views on education. The reason has to do with his way of fusing philosophy and social theory. He is the owner of an upto-date general store, a modern social philosopher of an old-fashioned sort. I do not mean
that his goods are old-fashioned. Indeed, he purveys a solvent for the very latest riddle in
action theory, as we shall see. I mean that, like many traditional political theorists, he hopes
that the study of men as they are will yield a science of laws as they should be. Accordingly, our theme will be his account of human nature, social action and freedom. Having
begun with a gesture to Hegel and a contrast between Humeans and pragmatists, I shall
introduce William Jamess search for a basic principle of personal unity and Deweys
offer to find it in the idea that man is a creature of habit. A theory of freedom follows and
its modern interest will be shown by considering Deweys two notions of motive, which
yield a normative theory of the self-in-action to occupy the rest of the paper. Interestingly,
it is a theory of self-in-action without a self and there will be complaints from philosophers
The self in action 37

and sociologists. Yet neither party is in a position to cast the first stone and perhaps neither needs a self, if they will only cooperate. I shall indeed urge them to co-operate under
Deweys guidance, in hope of a new notion of identity and a rational social policy. But they
will still need more of a self than Dewey provides, if freedom is, as Dewey has it, a matter
of forming institutions to let men translate into social action their essential human nature.
In the Festschrift for Deweys eightieth birthday Gordon Allport complains in a perceptive essay that Dewey can offer no proper theory of personality and Dewey, replying at the
end of the volume, more or less agrees.2 In the end, as I shall try to show, he must. Yet he
need not concede easily, since his theory of habit will do almost all he asks of it. Man
he maintained, is a creature of habit, not of reason nor yet of instinct.3 It is a deceptively
simple basis but it supports an entire analysis of the self, of social institutions and of human
freedom.
We shall do well to remember that Dewey had a Hegelian upbringing. He never threw
it off and it remains a distinctive thread in an American pragmatist who saw himself as
a good empiricist. But it was not a handicap of the sort which afflicts the socialist born
a viscount or the Jew once a gentile, for it gave him a sense that objects are always in a
state of becoming, a sense which is the stuff of pragmatism. Admittedly he abandoned his
early efforts to psychologize in a Hegelian mode. His Psychology of 1886, for instance,
employed a self, described as a real unity, to account for the connectedness and peculiarity of human experience. It was marked by an idealist belief in the dynamic union, indeed
total integration, of experience and nature. Later works, by contrast, treat integration as a
rare and precarious construct. Experience is no longer taken as essentially cognitive; feelings and activities are deemed at least as important. Nature ceases to be a single cosmos
and its purposive elements shrink to the organic subjects of human sciences like biology. The self drops out of obvious sight, along with its props of will and cosmic impulse.
Hegel had yielded to a twentieth-century, matter-of-fact view of man and naturewitness
Deweys wry remark in 1910 that he knew of a college so backward that it was still using
his Psychology as a textbook.
The Dewey we recognize is a twentieth-century figure who sees no single or rational
system in the world and who makes human freedom a fragile creature of education. But
Hegel has something to offer a pragmatist, and Dewey in any case was never a high and
mighty Hegelian on first name terms with. the Absolute. Even in Psychology his accent is
on process or becoming and the sentence referring to the self as a real unity reads in full:
The self is a connecting, relating activity and hence a real unity, one which unites into
whole all the various elements and members of our knowledge (part I, chapter 9, section
II). This bullfrog of a self asserts more than it explains and Dewey grew dissatisfied with
it for that reason. But, although he dropped it from obvious sight, the problem it pointed to
continued to vex him and should, I dare say, vex recent pragmatists too.
The problem is one of personal and social identity. It confronts Dewey through his
endeavour to work an instrumental theory of knowledge, a view of reality as process and an
account of the mind as connected activity into an empiricism descended from Hume. For
Hume, experience is given, the knowing mind is passive and there are no links in nature.
Knowledge of the world is therefore referred to perception and habits of association in the
mind. Even the relation of cause and effect, which is proclaimed as our sole way of going
beyond mere ideas and the data of the senses, reduces, from the standpoint of knowledge, to

concomitance. Admittedly Hume allows us imagination and assigns it some role in forming
our beliefs. But there is no rational warrant for the work of imagination and it has no place
in justifying claims to knowledge. Equally the self, whose ancestor is visibly Descartess
ego, turns out to be a mere bundle of passive perceptions. Hume had hoped for a science
of man and society from these minimal ingredients but, as he confessed so gracefully in
the appendix to the Treatise, they did not give him enough. He could not finally explain
the principles which unite our successive perceptions in our thought or consciousness. The
identity of the self could be neither explained nor explained away.
The last paragraph may be unfair to Hume but it makes a pragmatist line easy to spot.
We start by making the mind active instead of passive and make knowledge of the world a
matter of construction instead of perception. The minds job is not to attend to experience
in order to discern patterns but to work patterns into experience with the aid of imagination. There are no brute, atomic, uninterpreted facts; judgment, the applying of concepts
to experience, occurs a stage earlier than empiricists have been wont to believe. Judgment occurs at the moment of presentation, with the result that nothing is given and the
old strategy of erecting a temple of knowledge on certain foundations is to be rejected.
Instead of a temple we have a web of beliefs, spun and adjusted by the active mind. The
web constitutes knowledge because, I suppose, it satisfies us, perhaps for some reason of
principle like economy or usefulness, perhaps because of our human constitution. The
theory of knowledge is fused with the philosophy of science to yield a philosophy akin to
instrumentalismor, if it does not, Dewey will be entitled to complain.
In switching from a passive to an active mind, pragmatism sets itself a hard question
about how to relate experience, regarded as an interpretation, to experience regarded as a
test for beliefs. The latter seems to require givens which the former seems to deny. This
is an old problem which remains a centre of dispute. But there is an aspect of the dispute
which, from oversight or embarrassment, has been less prominent of late. Even if a passive
mind can be passed off as a bundle of perceptions, an active mind is pivotal. In conceiving
of knowledge as a web which is woven and rewoven, pragmatism does indeed pose queries
about the criteria of design and adjustment. But it poses no less obvious or urgent queries
about the weavers. James, Dewey and Mead were brave enough to tackle the topic but they
hardly disposed of it. If knowledge and society are to be seen as processes, what shall we
say of the spiders in the web, the spinners of the fabric, in a word of the self?
Although we are addressing ourselves to Dewey, we shall, like him, need to lean on William James. Dewey acknowledged his debt to Jamess justly famous views and dissented
from them only enough to make life easy for the social theory of G.H.Mead. James had distinguished between the I and the Me, dividing the latter or phenomenal self into the bodily
me, the social me and the spiritual me.4 His account of the Me is riveting enough but we
must restrict ourselves here to the I. Many a pragmatist would brush the I aside but James,
bless him, did not. Admittedly he had no patience with substance theories, since to say
that phenomena inhere in a Substance is only to record ones protest against the notion that
the bare existence of the phenomena is the total truth (p. 346). But he does indeed record
his protest and is no less impatient of Humean, associationist attempts to treat the self as a
bundle of self-adhesive perceptual atoms. Transcendentalism, he says next, has the merit
of recognizing the problem but serves only to pose it. What, then, is the basic principle of
personal unity? James concludes that the self, the I, is a present Thought which remembers

and so appropriates past members of the stream of consciousness of which it is the present
unit. This, he hopes, will account for the unity of the self and the difference between one
self and another, without entering the realm of baseless speculation. The Me is the empirical person and the I the judging Thought.
Dewey admired and adopted the general approach but changed the emphasis. Jamess
view could be deemed intellec-tualist, a case of cogitat ergo sum, so to speak. For Dewey
there is no such thing as pure thought or Thought. Thought presupposes habit and involves
action. Thinking, like all other intelligent activity, requires having already acquired an
ability. An ability is a habit and its exercise depends on social relations. All life operates
through a mechanism and the higher the form of life the more flexible the mechanism
(HNC, p. 70). Possession shapes and consolidates the I of philosophers. I own therefore I am expresses a truer psychology than the Cartesian I think therefore I am (HNC,
p. 116). Dewey links James to Mead, by taking the I as the Thought in social action, imposing order on the experience of social life and at the same time adapting to it. As promised,
the linking concept is that of habit.
For Dewey, man is a creature of habit. This sounds like the kernel of a stimulus-response
theory, with Skinner-like implications for science, ethics and education. But Dewey has no
such thing in mind. Habit is not just conditioned response but energy organised in certain
channels (HNC, p. 76). Nor is he writing a bores charter, even though the nature of habit
is to be assertive, insistent, self-perpetuating (HNC, p. 58). In place of habit as passive
routine, he offers us habits as ways of translating desire into intelligent action. All habits
are demands for a certain kind of activity; they constitute the self (HNC, p. 25). There is, in
short, something inside the black box with energy and power to organize experience. The
free man is the man who uses the power aright. The key to freedom is a simple but striking
idea. The free man is the man of rational habits.
The phrase rational habits wears an air of paradox. It is easy to think of rational action
as action which is deliberately chosen under the conscious influence of good reasons.
Equally, habitual actions are easily thought of as those done from mindless routine. So it
is easy to oppose the rational and the habitual, with the result that large areas of life are
transferred to the realm of the non-rational. I do not, for instance, drive my car rationally,
if the test is that I must choose each and every action by reflecting consciously on what I
should do next. This line of thought has its temptations. It has tempted the sociologist to
argue that, since so little of life is rational, he can ignore the notion of rationality in studying social action. It has tempted the moralist to urge us that we can be rational and free only
if we become always critically self-aware. The common theme is that creatures of habit
cannot be creatures of reason.
The snag is that a fully rational man, as just defined, would never get as far as his own
front door. Each morning would find him barefoot in his bedroom, trying to decide which
sock to put on first. Were others equally rational, civilization would collapse into paralysis,
like some giant centipede told to put its best foot forward first. The error is to forget that
freedom depends on skill and skill on habit. Motorist and mountaineer, call-girl and diplomat, yogi and commissar, juggler and philosopher can pursue their own good in their own
way only if they are creatures of habit. The motorist literally and the others metaphorically
need an unthinking control of their vehicles.

Such skills become, in the eyes of James or Dewey, parts of the self. But they do not
make the self passive. They extend its activity and so extend the self. Yet this is not flatly
true, since there are also deadening routines which stunt the self. What is necessary is that
habits be formed which are more intelligent, more sensitively percipient, more informed
with foresight, more aware of what they are about, more direct and sincere, more flexibly
responsive than those now current (HNC, p. 128). Man is a creature of habit but there are
two kinds, not of habit, but of men. In some of us habits are divided against one another,
personality is disrupted, the scheme of conduct is confused and disintegrated. In others
released impulses are intelligently employed to form harmonious habits adapted to one
another in a new situation. There is a nasty question here about the criteria for deciding
when a habit is disruptive and to be overcome or when it is alerting and to be welcomed.
Dewey is teasingly vague, as we shall complain later. Nevertheless, in so far as he can characterize good habits, the moral for education and for freedom stares us in the face.
Dewey takes the theoretical question about freedom a step further in a splendid essay
entitled Philosophies of freedom.5 The essential problem of freedom, he says, is that of
the relation of choice to unimpeded effective action. The key is that choice is not mere
impulse, whim or will; nor is effective action mere power to do, as it is for an avalanche
or earthquake. Choice signifies a capacity for deliberately changing preferences (p. 276).
It presents itself as one preference among and out of preferences; not in the sense of one
preference already made and stronger than others but the formation of a new preference
out of a conflict of preferences (p. 276). Effective action is the enacting of rational preferences and freedom lies in the development of preferences into intelligent choices (p. 296).
Freedom is thus not the impulsive movement of an abstract, pre-social individual towards a
given goal. It is the intelligent harmonizing of habits for the open-ended purpose of becoming an intelligent human being.
Why is this approach of living interest? The reason is epitomized by the strong actionist
trends in social theory, bearing labels like phenomenology or ethnomethodology. Their root
idea is that men do, or at least can, create their own social reality out of shared experience
by acts of self-definition. It is not wholly a novel thought. Theories of the social contract
made society the work of individuals. Notions of subjective and intersubjective meaning
have long been the stock in trade of symbolic interactionism. Webers use of meaning and
action to explain social forms is as central to one grand line in sociology as Durkheims
emphasis on structures is to another. Recently, however, the idea of reality as a construct
has become the rallying cry for a radical or humane approach to the social sciences, which
rejects all these forms of positivism in fields as different as history, physics, criminology
or education. Although positivism is a murky term in this context, I suppose it is meant
to embrace all attempts to find law-like determinants of action external to human agentsan-sich. The villains are not only those, like Durkheim, who look directly to external and
constraining social facts. They are also those who start with inner meanings but either take
them as given and their construction as unproblematic or go on to explain them by appeal
to social structures, thus merely lengthening the causal chain between structure and action.
Weber is deemed finally as misguided as Durkheim. Homo sociologicus falls but the crown
does not go to homo psychologicus, for psychology, too, has been a generalizing subject in
search of law-like determinants of action, whereas strong actionism demands a theory to
make human actors individual and autonomous.

Men are to be brought back in to explain the construction of reality. But what sort
of men and how does their existence explain? By rejecting law-like determinants, strong
actionism seems to abandon hope of ever explaining anything; and, by pouring scorn on
the atoms of old liberalism and methodological individualism, seems to forfeit all chance
of a self to sustain the construction of reality. Strong actionism, in short, threatens to be so
radical that it dashes all hopes and has to end by borrowing eclectically from its opponents
and hiding its debts in bombast. To escape this by now usual fate, it must take the I of the
I-and-the-Me seriously, must produce an I discernibly sociological and must find a mode
of explanation not law-like but still explanatory. Not even the kindest critic could suppose
that it has yet succeeded.
Deweys writings are therefore topical. He does indeed promise a coherent theory of
the self in social action which is genuinely explanatory. But we should treat him gingerly.
His mind runs on an intoxicating fluid which, as a recent advertisement for lager puts it,
refreshes the parts which ordinary beers cannot reach. This gives it high performance and
huge output but makes for a certain casualness. For instance, we are told on page 42 of
Human Nature and Conduct that habit means will, and on page 52 that will means habit.
Such circularities are common and precision in one place is all too often cancelled by an
embracing vagueness elsewhere. None the less the promise of a grand theory is there and I
shall try to extract it by seeing what Dewey has to say about motive.
The ingredients of action, let us assume, are an environment, an occasion, an agent
and an outcome. For instance, during a battle the arrival of enemy tanks sets the general a
problem, which he solves by ordering his men back to the hills. What explains the generals
action? That, no doubt, is an empirical question. But it is one only in so far as we already
have a theory of action and motive which makes it one. Theories which perform this service can be grouped typically into the adaptive and the constructive. By an adaptive theory
I mean one which treats the action as a conditioned response to the occasion; by a constructive theory one which takes the action as a step in the execution of a policy. Stimulusresponse models are clear examples of the adaptive but they are not the only ones, nor are
all adaptive theories behaviourist. Room can certainly be made for subjective and shared
meanings. Allowance can be made for how the agent perceives his situation. Indeed, there
is a place for the strong thesis that the very identities of the occasion and outcome depend
on the description which the agent would give of them. The softer brands of actionism differ greatly from mechanical models. But, in echo of the radical critique of Weber, they, too,
rely finally on causal determinants of action, usually beyond the realm of meanings and
certainly beyond the actor-an-sich. This is, indeed, the litmus test for dividing actionisms
into strong and weak and it places softer brands, like Webers, in the adaptive category, for
adaptive theories are those which try to explain a tergo with the aid of general laws. They
are in one endlessly disputable word, determinist.
Dewey firmly denies that motives are causes working a tergo (HNC, p. 120):
A motive does not exist prior to an act and produce it. It is an act plus a judgement on some
element of it, the judgement being made in the light of the consequences of the act.
Instead then of saying that a man requires a motive in order to induce him to act, we should
say that when a man is going to act he needs to know what he is going to dowhat the
quality of the act is in terms of consequences to follow. A motive in short is simply an
impulse viewed as a constituent in a habit, a factor in a disposition.

As so often, he gives us a broad idea of what he has in mind, while making it hard for us to
pin him down. Broadly, questions of motive are to be addressed to acts as a whole, in order
to elicit the actors habits and judgment of consequences. But what precisely might this
come to for the general and the enemy tanks? The general is no doubt recording his bold
judgment of tactics for the sake of his memoirs. His batman is perhaps noting an impulse
of fear as a constituent in the generals habit of cowardice. Both are following Deweys
instructions but they disagree not only about the actual motive but also about the sort of
motive. What is Dewey to say?
The masters reply cannot be wholly satisfactory, given the casual way in which he
makes disparate ideas share the same umbrella, but he is not left speechless. As in his
account of freedom, he leaves the truth to turn on the degree to which the generals habits
are intelligent. Impulse is an intermediary. It offers imagination and invention a chance,
which they may or may not take. Impulse is needed to provoke thought but thought, born
as the twin of impulse in every moment of impeded habit (HNC, p. 171), can master
impulse. If the general is a thoughtful man, he takes his chance and chooses freely by forming a new preference out of a conflict of preferences. If not, then the batmans explanation
is the true one. In effect, therefore, Dewey has two incompatible theories of motive and
resolves the conflict by giving each a different range of application. A motive is sometimes
an impulse viewed as a constituent in a habit, sometimes an act plus a judgment on an element of the act. The former is the motive of an unenlightened man, the latter of a free man,
and the implications Dewey draws for ethics, politics, education and democracy will be
known to all his readers.
If Dewey rejects adaptive theories of motive for free action, does he then accept
constructive ones? That depends on the further question of how far the explanation
of action as a step in the execution of a policy involves postulating a self. In saying
that the general pulled his troops back as a tactical means to a strategic end, what is
assumed about the general? Hobbes defines freedom as the absence of all impediments to action which are not contained in the nature and intrinsical quality of the
agent and this is a question about the nature and intrinsical quality of the general.
Dewey scorns a substantive self for William Jamess reasons and does not believe
that human conduct displays the rational unity which is sometimes made the sign of a
self. The truth is that there is no one ready-made self behind activities. But there is
still a self, the point being only that it is neither one nor readymade. We arrive at the
conceptions of motivation and interest only by the recognition that self-hood (except
as it has encased itself in a shell of routine) is in process of making, and that any self
is capable of including within itself a number of inconsistent selves, of unharmonised
dispositions (HNC, p. 137). Once again the key lies in habits and in the relation of
habits to impulses (HNC, pp. 177ff.):
Concrete habits do all the perceiving, imagining, recalling, judging, conceiving and reasoning. Yet habit does not, of itself, know, for it does not of itself stop to think, observe or
remember. Neither does impulse of itself engage in reflection or contemplation. It just lets
go. Without habit there is only irritation and confused hesitation. With habit alone there
is a machine-like repetition, a duplicating recurrence of old acts. With conflict of habits and
release of impulse there is conscious search.

The crucial phrase here is conscious search. Those who reject an adaptive theory of action
with its attempt to find determinants a tergo often propose a rational man model instead.
This rational man has clear goals and his actions are explained as rationally calculated
means to his goals. If that is what constructive theories of action are about and if that is
what explaining action as a step in the execution of a policy comes to, then Dewey rejects
them too. An intelligent human being is consciously searching not only for means but also
for goals. He is always in a state of becoming and the conscious search is a search for himself. The search is not a treasure hunt, however, since there is no pre-existent self to find.
The search creates the self by bringing habits into harmony, thus translating the Thought
into social action and constructing a man who is free and human.
The view is attractive but teasing and, having only a short time, I shall focus on the
feature which most attracts and teases me. It is its notion of personal identity, designed
to tempt philosophers and sociologists together. Such a virtue is rare, if not impossible.
For philosophers discussing personal identity are far removed from sociologists discussing social identity and it is easy to suppose that wholly different concepts of identity are
involved. Dewey would maintain, I fancy, that there is none the less a unitary concept to
be had and would invite us to discern it in his translation of Jamess Thought into social
action. Both parties are likely to object. Sociologists will complain that Deweys free man,
constantly seeking a harmonious self with no fixed goals or criteria of harmony, is chronically prone to anomie; that, where there is chronic anomie, there can be neither identity
nor explanation. Philosophers will complain that Dewey has no notion of strict, numerical
identity, without which any theory of personal identity falls. These are dissimilar complaints yet they share a demand for a fixed point in a qualitative world. If Dewey can meet
the demand or show it importantly misguided, both parties will learn something to their
advantage. My own view is that there ought to be a unitary concept, which would harness
the two disciplines. Without going so far as to say what it is, I shall praise Deweys attempt
to find a harness and suggest what it lacks. To do more would only reveal my own ignorance and I offer criticism in no spirit of lordly scorn.
Philosophers and sociologists certainly seem to use different senses of identity. The
former typically seek logical and epistemological criteria by which persons can be identified and re-identified. The latter typically look for a theory of personality which will fuse
psychology with role theory. It would be only a poor pun to say that both were interested
in relations of identity, since logical relations like transitivity belong in a different universe
from social relations like role partnership. There appears, in short, to be nothing of a philosophical interest which a crisis of identity is a crisis of. At the same time, however, neither
discipline can be said to have emerged triumphant from its peculiar inquiries and there is
at least this excuse for suggesting an alliance.
Of the philosophers stock questions, I pick two: What is a person? and What unites
different stages of the same person? As they are old questions, possible answers must
pass old, severe tests. Essential attributes of a person and logical guarantees of continuing
identity are demanded. For instance, if disembodied existence is possible, I cannot be my
body, however sure we are that all persons are in fact embodied. Memory is not a criterion,
if two different people could possibly have the same memories, however well we know that
it does not happen. Theories of identity are tested to destruction by Gedankenexperimenten
and there are at present no undisputed survivors. What is a person? The concept of a person

may be a primitive in relation to those of body and experience but we can hardly leave it
at that and there is no easy riposte to Jamess attack on substantive, associationist or transcendental theories. Those who think persons essentially physical cannot yet specify the
predicates which distinguish persons from other bodies. Those who think them essentially
non-physical, cannot yet find a way to individuate consciousnesses. Similar (and, you may
protest, equally tendentious) strictures apply to attempts at the other question. What unites
different stages of the same person? Bodily continuity, however necessary, does not seem
sufficient; non-physical criteria, however subtly they take account of mental life, do not
seem to secure uniqueness. The hardy perennials can be posed in a minute and explored in
an hour without making much advance on centuries of bafflement.
How tempting it is, then, to invoke some idea of social relations. Consciousness does
not operate in a vacuum and bodies are just bodies unless the relations between them are
endowed with shared meanings. James remarks: Properly speaking, a man has as many
social selves as there are individuals who recognise him and adds that, more loosely, he
has as many social selves as there are distinct groups about whose opinion he cares (Principles of Psychology, p. 294, his italics). If we take the hint and look to sociology, we shall
find at least two interesting lines of thought. One lies in structural role theory, where the
self is externalized as the significant social behaviour of the individual and then absorbed
into the roles attached to his social positions. The other follows the subjective path from
symbolic interactionism until the self becomes images seen in the eyes of others, reflected
to infinity. The snag from the philosophers point of view, however, is that these lines do
not yield notions of strict identity. A social position can by definition have more than one
incumbent, so that each of us, even if in fact uniquely placed in some social system, is not
guaranteed uniqueness by citing the sum of his roles. Equally a person (or sort of social
Leibnizian monad) defined by means of reflections in the eyes of others is logically unique,
only if (pace Leibniz) some person not so defined is used as a fixed point of reference. Most
philosophers will deem it neither essential to my being a person that I have social positions
nor logically necessary to my being the person I am that my social relations are in sum
unique to myself. The sociologist can elaborate until he is blue in the face, by adding to my
role set the role sets of my role partners and of my role partners role partners, throwing
in the historical order in which I played my roles, listing my most significant experiences
and even sneaking in mention of genotypes and phenotypes. These are all factors in my
conduct and an intimate part of being me. But the philosopher is implacable. He demands
strict identity, while the sociologist supplies something too lax to be identity at all.
Yet perhaps the philosopher is asking too much or the sociologist trying too little. In
support of the latter alternative, there is a classic lacuna in role theory just where we might
expect to learn who or what plays the roles. Readers of Erving Goffman, for instance, relish his nuanced accounts of how actors play their roles with varying styles and skill, at
varying distance and for varying ends. But, noting that the social world is thus made to
depend largely on the man in the masks, they look in vain for a theory of the self. To say,
as Goffman does in The Presentation of Self in Everyday Life, that a self is a repertoire of
behaviour appropriate to a given set of contingencies is, however prettily epigrammatic, to
speak in circles: for what makes the behaviour appropriate (or even a repertoire at all) is the
ultimate character of the self and other selves, which the behaviour is being used to define.
The same odd lacuna occurs among phenomenologists and others who would construct the

social world from acts of self-definition expressed in shared meanings. Here an epistemological theory of the self is surely crucial, if individual men are to be justifiably presented
as the creative point dappui for the analysis of an intersubjective reality. But, apart from
some hard sayings by Husserl, which his Anglo-Saxon descendants usually reject, none
springs to mind. Perhaps the gap could be filled, if the philosophers demand for a notion
of strict personal identity were taken seriously.
James and Dewey would take the other option, that, since reality is process, the philosopher is asking too much. There cannot be strict identity between distinct members of a
series, and pragmatists, like Humeans, can relax. The Buddhist image of the self as a series
of candles lit from the stub of the one before appealed to James. He analyses a person as
a remembering Thought and sets about the problem of personal identity by combining
philosophy with psychology. Dewey, as we have seen, works in the same spirit but offers
his idea of self-monitoring habits as the key to a door into sociology unlocked by Mead.
Jamess I and Me, allied with Deweys instrumentalism, tempt the stern philosopher to
rethink his strict criteria of personal identity. Perhaps there is virtue in yielding. Indeed
there certainly is, if A.J.Ayer is right in remarking:6
If one speaks of the construction of objects out of the flux of experience, it is indeed natural
to ask who does the constructing; and then it would appear that whatever self is chosen for
this role must stand outside the construction; it would be contradictory to suppose that it
constructed itself. But the metaphor of construction is here misleading. What is in question
is the derivation of concepts, not the fabrication of the things to which the concepts apply.
To construct either the material or the spiritual self is to do no more than pick out the relations within experiences which make it possible for the concept of a self of this kind to be
satisfied, and these relations exist whether or not we direct our attention to them.
The issue is too hard to settle at a stroke but I can at least open a line of thought. It is
prompted by the reflection that a theory of human identity determines what should count as
good reasons for action. I am not speaking of means-ends rationality with given ends and
calculable means; nor of good reasons for pursuing ends which are in turn means to further
ends. My question concerns good reasons for ultimate ends. The traditional answer is that
we have ultimate interests deriving from our essential human nature. Hobbes and Rousseau, for instance, were not simply issuing prescriptions off the tops of their heads with
wild leaps from is to ought. They hold that whatever constitutes us as human beings is
eodem ipso what we should preserve and develop. Their analysis of what is, no less than of
what ought to be, rests on essential attributes of mankind. Such a strategy is out of fashion
and is one reason often given for holding that traditional political theory is dead and buried.
But, although buried, it is far from dead; for, even if human nature is specific to economic
formations, modes of operant conditioning or given forms of life, these are still universal
judgments about mankind. Even if a person is a body or a process with only qualitative
identity, this is still a metaphysical thesis, to be judged, I would argue, by its claims to truth.
Dewey, at any rate, holds both that his analysis of human nature is true and that it requires
a democratic form of society, for democracy alone lets us cultivate rational habits and only
by living a life of rational habits can we be truly human. Problems of personal identity are,
for him, questions in what we might call the political philosophy of mind. Human Nature
and Conduct is no careless title.

But his strength is also his downfall. The strength is that he works without given goals
and refuses to talk the language of means and ends. The self is constituted by its habits,
habits are malleable, means and ends are not distinguishable, criteria for guiding action are
fluid and the self is, in short, always in a state of becoming. This has interesting implications both for the philosophy of mind and for the utilitarian social sciences, like neo-Classical economics. But there is a crucial snag. Rational habits are those more intelligent, more
sensitively percipientmore flexibly responsive than those now current (HNC, p. 128).
If this is, in upshot, all Dewey can say, he faces an evident circularity. Rational habits are
those which spread harmony among a mans discrepant selves; selves are discrepant when
preferences are not turned into intelligent choices and so into effective action; choices are
intelligent and action effective, when they let the self develop; the self develops when habits become more rational. There can be no exit and the charge of chronic anomie is justified.
But although it is true that, where there is chronic anomie, there can be neither identity nor
explanation, this is less a sociological fact than a philosophical consequence of having no
independent criterion of harmony.
If Deweys embarrassment is not total, it is because he is less noncommittal about ends
than he makes out. On close inspection, he is seen to advance only the stock Enlightenment
view that men have no goals outside themselves. He still assigns them the goal of being
human.7 It is a moving target, a more genial version of Hobbess connection between Felicity and our perpetual and restless desire of power after power, that ceaseth only in death.
But it is still a goal, translated by Deweys theory of freedom into reasons for action under
the spur of conflicting impulses. To this extent he is inconsistent. Nor will so vague an idea
of what it is to be an intelligent human being stave off the critic who asks why we should
pursue this ill-defined course. A fool satisfied is still a man and a satisfied man at that.
Turning a concept of a person into reasons for action is a hard alchemy and I do not
pretend to have the philosophers stone. But we can see what is needed, if we recall the
old notion of conatus, signifying the endeavour of each thing to persist in its own being.
According to Spinoza it is nothing else but the actual essence of the thing in question (Ethics, III, props VI, VII) and can be used to analyse resistance in passive things, freedom
in active ones. Action which impedes conatus has its causes outside the agent, whereas
rational action can be explained by the necessities of the agents own nature. A true theory
of essential human interests would apply descriptively to the conduct of free men and prescriptively to all. It would solve the old problem of identity by showing what was essential
for each things being this particular individual member of its species. Then perhaps we
would know which habits to cultivate, why only a fool is content to be satisfied and even
which theory of education to apply.
While waiting for the philosophers stone, we could do worse than read Spinoza. But
that is not why I mention him. He appears here only to score a point about the rationality of
ends. A system as elastic as Deweys leaves no possible test for intelligent and harmonious
habits. The trouble stems from an elastic criterion of personal identity. That is, I submit,
one reason why the self must stand outside the construction and why James was right to
accept the need for a basic principle of personal unity, while rejecting Kants claims to
have found one. In final comment on Deweys Encyclopedia article, with which we began,
there have to be some limits to the modifiability of human nature, since we could otherwise
have no reasons for action, and some limits to possible institutional change, since a change

which destroys the identity of men in society also destroys all institutions. From these limits we can hope to derive the outline of a rational social policy. Experimental observation
continues to have its vital place but only within an a priori theory which sets its scope, with
due respect for uncertainty, given our theoretical ignorance.
Yet nothing in this conclusion favours the shopping centre against the general store.
Sociologists who repair to Dewey will find a bargain analysis of rationality and its use in
explaining the self in social action; philosophers will be tempted by theories of motive and
human identity which make social relations relevant to an old enigma. These goods may
not work perfectly but there are few and shoddier rivals on sale in the specialist shops up
the road.
Notes
1 I am grateful to the John Dewey Foundation for instigating what has proved an absorbing task and
to Bryan Heading and Quentin Skinner for their kind help and criticism.
2 Gordon W.Allport, Deweys individual and social psychology, in P.A.Schlipp (ed.), The Philosophy of John Dewey, Tudor, New York, 1939. For Deweys comments see esp. pp. 555f.
3 Human Nature and Conduct, part II, section IV, p. 125. This being the central text I am relying
on, I shall refer to it as HNC. Page references are to the Modern Library edition, Henry Holt, New
York, 1930.
4 These names for the divisions are the ones used in Psychology: Briefer Course, the shorter version of the theory first put forward in The Principles of Psychology, Henry Holt, 1890, ch. 10. My
next page reference is to this chapter of The Principles of Psychology, Dover edn, 1950.
5 Originally in H.M.Kallen (ed.), Freedom in the Modern World, Coward McCann, NewYork, 1928,
pp. 23671; reprinted in John Dewey, Philosophy and Civilisation, Minton Balch, New York,
1931, to which my page numbers refer.
6 The Origins of Pragmatism, Macmillan, 1968, part II, ch. 3, p. 261. Ayer lends distinguished support to James by working out an elegant qualitative theory of personal identity. There is also the
interesting case put by Derek Parfit in Personal identity, Philosophical Review, vol. LXXX, no.
1, 1971.
7 See, for instance, Philosophies of freedom, reprinted in Philosophy and Civilisation, Minton
Balch, New York, 1931, pp. 286ff. especially p. 287.
5
Democracy and education
Antony Flew
Perhaps my title ought to have been given in italics, for I shall in the event be concerned
mainly with the subject of democracy and education as, and in so far as, this is treated in one
particular Dewey book.1 But he himself, in an autobiographical fragment, described that
book as having been for many years the source in which, not merely his ideal of education
for democracy, but his entire philosophy, was most fully expounded.2 My treatmentlike
Caesars Gaulfalls into three parts. The first distinguishes at some length, and without
direct reference to Dewey, three established areas of meaning for the word democracy.
The second, and by far the most substantial, examines Deweys two proposed criteria for
an ideal society; which would, he affirms, be democratic. The third and shortest notices
one or two educational recommendations not very closely tied to the central theme of this
particular book. All three parts, and the notes thereto, make various more or less unfriendly
references to Education for Democracy,3 a widely circulated and, presumably, influential
work by a group of Penguin Educationalists. The relevance of these references is as links
between Dewey and current debate.
1 Preliminary inquiries about democracy

Anyone who once long ago sat at the feet of the implacable professor, and who nevertheless undertakes now to read a lecture entitled Democracy and Education, must begin with
some apology; for did not John Austin give democracy as his paradigm case of a notoriously useless word?4 Certainly it is both ambiguous and vague. Yet on neither of these two
different counts can it be abandoned as futile. I shall in a moment myself be urging that we
have to distinguish at least three main areas of meaning. But that a word has two senses,
or more, is never by itself sufficient reason to conclude that it is even desirable, much
less practical, in one or more of these present senses to replace it by another. Nor would
it be sensible to propose that we should, especially in our talk about matters of the highest
importance, dispense with all words which are in any direction vague.
At first hearing that last suggestion may indeed appear obviously sensible. Yet a short
pause for thought will reveal that it must be utterly impossible to do without either precise
expressions or vague expressions. We have to have ways of saying, for instance, both four
oclock on the dot and some time in the middle of the afternoon. More immediately to
the present point, many of the most important of all differences happen to be differences
of degree. They are, that is to say, differences in which the clear-cut paradigm cases at one
extreme are linked to the clear-cut paradigm cases at the opposite extreme by a spectrum of
Democracy and education 49

actual or possible, more or less marginal, cases in the middle. Until and unless we choose
to draw some arbitrary line, therefore, the terms employed to characterize such opposites
cannot but be vague: To remove vagueness is to outline the penumbra of a shadow: the
line is there after we have drawn it; but not before.5
Yet nothing could be more wrong, nor more common, than to assume, or even to say, that
all differences of degree are mere differences of degree, for, defined as we have just defined
them, these supposed mere differences of degree include many if not most humanly vital
differences: the differences, for example, between youth and age; between riches and poverty; between sanity and insanity; and between a free society and one in whichas used to
be said of Imperial Germanyeverything which is not forbidden is compulsory.
(a) By the people

In the first of the three areas of meaning to be distinguished the word democratic is
applied to methods of making state or, more generally, group decisions. If some group as
a whole takes decisions by majority vote, then that is democratic. So, too, are institutions
under which decisions are made by delegates, representatives, or other officers who can in
due course be voted out.
Two things in this indication of the first area of meaning require further explanation in
the present context. The first is my emphasis on voting out, not voting in. For todays world
this is of great topical importance. In the newly created states of formerly British Africa, for
instance, most of the original regimes were established as the result of tolerably presentable
elections. But nearly all of these regimes then jettisoned their claims to democratic legitimacy by proceeding in fairly short order to take steps to ensure that it should be impossiblein that good old phraseto vote the scoundrels out. The second thing to notice is my
studiously nonpaternalist emphasis upon what voters do actually want; as against what they
may be supposed to need, or what it may or may not be in their interests to have.
(b) For the people

The importance of that second emphasis emerges more fully as we move to the second area
of meaning. Consider such increasingly common political labels as Peoples Democratic
Republic of the Yemen, Democratic Republic of North Vietnam, or Somali Democratic
Republic. Those of us who still remain devoted to democracy as first conceived may too
quickly put down such employments of the hooray word democratic as just so many more
specimens of sickeningly mendacious propaganda. But that is not the whole story, and the
other part is for us the more important.
Two authoritative statements will bring out what that other part is. My first witness is
Janos Kadar, addressing the Hungarian National Assembly on 11 May 1957, one year after
the friendly neighbourhood Soviet tanks installed him in office:6
The task of the leaders is not to put into effect the wishes and will of the masses. The task
of the leaders is to accomplish the interests of the masses. Why do I differentiate between
the will and the interests of the masses? In the recent past we have encountered the phenomenon of certain categories of workers acting against their interests.

As my second witness I call the late Abdul Kharume, First Vice-President of Tanzania. Mr
Kharume, who has since been assassinated, was, as his Afro-Shirazi Party in Zanzibar still
is, strongly influenced by advisers from the German Democratic Republic. Referring to a
recent round-up of the unemployed in Dar-es-Salaam, he said:7
Our government is democratic because it makes its decisions in the interests of, and for the
benefit of, the people. I wonder why men who are unemployed are surprised and resentful
at the Governmentsending them back to the land for their own advantage.
We can, therefore, distinguish a second political sense of the word democratic. It is, presumably, derivative from the first. The crux is suggested by Janos Kadar, and spelt out by
Abdul Kharume.
(Parenthetically: those interested in the history of ideas will see Rousseau as, whether
willy or nilly, one of the main intellectual forebears of this second concept of democracy;
for Rousseaus often deceived but never corrupted General Will is always upright, and necessarily directed to the collective good, while the notorious fact that it is not to be reliably
discovered, either in the hurly-burly of contested elections, or through the deliberations of
representative assemblies, gives such a doctrine strong appeal to all who think of themselves as belonging to a party of the vanguard.)8
(c) Of the people

The third area of meaning has, so far as I can see, no essential connection with the political.
The big Oxford English Dictionary glosses, with little enthusiasm: In modern use often
more vaguely, denoting a social state in which all have equal rights, without hereditary, or
arbitrary, differences of rank or privilege. Thus we may describe the social-life arrangements of some organization, which is perhaps in working hours hierarchical and authoritarian, as thoroughly democratic: the managing director plays football under the captaincy of
an apprentice; and so on.
Using the word democratic in this sort of sense, an institution which is in its recruitment
highly selective, and itself perhaps in many ways privileged, may properly be described as
democratic; providing only that recruitment to it is socially open, and not determined by
hereditary, or arbitrary, differences of rank or privilege. This granted it becomes urgently
topical to press upon all those who employ that currently most fashionable term of abuse,
elitist, the question: first, whether, for them, all selection and all segregation on the basis
of any form of achievement or capacity is elitist, and therefore bad; or, second, whether,
for them, the epithet elitist is pretty well synonymous with the word undemocratic
employed in a sense of this third kind. About this I will now say in passing only that, if
and in so far as someone is inclined to answer Yes to the first of these questions, then that
personthough in the darkness of these times, probably accepted as an educationalist
begins with that answer to take up arms against all education, all quality, all culture, and
all civilization.
There seems, as I have already suggested, to be no logically necessary connection
between the present third area of meaning for the word democratic and either of the first
two; although it is no doubt true that groups which are democratic in the first way, or even

in the second, will in fact tend to be so also in the third. Thus Alexis de Tocqueville has
a chapter with the prematurely Gaullist title: Democratic social conditions of the AngloAmericans.9 Yet he does not suggest that these conditions are more than contingently
connected with the possibility of upset electionsmuch less with the absolutism of a totalitarian socialist party of the vanguard.
The contributors to Education for Democracy are operating mainly in this third area;
although several are obviously committed to the first, and some to the second, sort of political democracy.10 For the thrust of the whole book is towards a universal, compulsory, comprehensive system; with the absolute minimum of either setting or streaming. Thus one of
the two editors concludes his own contribution: Without absorbing the independent, and
particularly the public schools, into the state system, any talk of education for democracy
remains a mockery (ED, p. 88).
2 Deweys concept of the ideal society

Let us now, a little late in the day, open our Dewey. The first sentence of his Preface
reads: The following pages embody an endeavour to detect and state the ideas implied in a
democratic society and to apply these ideas to the problems of the enterprise of education
(DE, p. iii). We might, therefore, expect forthwith to proceed to some explanation of what
Dewey proposes to mean by the word democracy; to be followed perhaps by an account
of the relations, or lack of relations, between this meaning and other established meanings.
In fact we do not. The first reference to democracy occurs in chapter 7. Nowhere is there
any treatment of other interpretations as such.
In chapter 7, under the title The democratic conception in education, Dewey seeks
a measure for the worth of any given mode of social life. The problem is to extract
the desirable traits of forms of community life which actually exist, and employ them to
criticize undesirable features and suggest improvement (DE, p. 83). He picks out two such
traits, happily discovering that these are precisely what characterize the democratically
constituted society (DE, p. 87). What they are is positive scores on the questions: How
numerous and varied are the interests which are consciously shared? How full and free is
the interplay with other forms of association? (DE, p. 83).
(a) The status of Deweys criteria

Sometimes the characteristics thus specified do seem to be being construed as if they were
elements in a definition of democracy. But these are occasions when Dewey is even further than usual from any consideration of possible principles and procedures of political
decision-making. Indeed he scarcely could have thought, when he did thus happen to have
our first area of meaning in mind, that there is any logically necessary connection between
that and these two most-favoured characteristics. The absolute maximum which could be
said on these lines is that his two criteria provide useful contingent indicationsas someone might suggest that we could arrange countries on a scale of the more or less easy for
their own citizens to get out of and for others to get into. But while we could perhaps do
something of this sort with Deweys second criterion, there would seem to be no comparable mileage to be got out of the first.

So I take it that we ought to consider these criteria, neither as tests for the presence of
democracy in any political sense, nor as contributions to the elucidation of such a sense,
but ratheras after all, Dewey himself suggested earlier on the same pageas offering a
comprehensive index of all the excellences of an ideal society (DE, p. 83). If there is any
necessary connection between Deweys ideal and democracy, then it will be not with the
first or second but the third area of meaning.
Two of the first things to remark about Deweys treatment, both in Democracy and
Education and elsewhere, are, therefore: first, how little he has to say about electoral politics; and, second, how, even when he does say that little, he never seems to notice that his
own usual understanding of the word democratic runs somewhere far from its primary
employment. The most striking but still partial exception which I have found to this general
rule is in the final chapter, Democracy and education in Schools of Tomorrow. The passage begins: Our famous brief definition of democracy, as government of the people, by
the people, for the people. It continues for many pages in that understanding.11
The same first two remarks apply with almost equal force to Education for Democracy. Several contributors do in one way or another reveal personal political commitments
either to participatory democracy or to democratic centralism. But almost no attempt
is made there to relate educational proposals to such political understandings of the word.
Neither do the authors show much appreciation of any need to explain their own present
preferred usage, nor to relate it to current alternatives. Nanette Whitebread, for instance,
presses her boo and hooray stops with undiscerning and uncritical abandon: Government
vacillationhas reflected the struggle between democracy and elitism, in which the latter
has been aided by pressure to restrict spending in a predominantly private, capitalist economy; and, Successive governments have been guilty of anti-democratic mismanagement
of teacher supply (ED, p. 176).
Since she, like most of her fellow contributors, typically contrasts democracy with elitism we can presume that she shares their primary present concern with our third area
of meaning. But, as neither she nor any of the others ever distinguishes senses, we are
disappointed of answers to our fundamental questions. For instance: what precisely is the
objection to independent schools? Is it really an objection not to independence but to nonacademic selectivity; an objection which could, therefore, be met by possible arrangements
to ensure that their admissions took no account of hereditary, or arbitrary, differences of
rank or privilege? Or is it an entirely different objection, essentially socialist, to independence as such? If we did get an answer it would be, I guess, that Nanette Whitebread
and most of her fellow Penguin Educationalists object on both grounds. They are, that is:
bothas socialistsagainst any independence from state or other collective public ownership and control; and alsoas a kind of educationalist, if not perhaps of educatoragainst
selection even on the most purely academic criteria. But these two objections are quite
different. They need to be separately stated, and distinguished.
(b) Formal rather than material criteria

Returning to Deweys two measures for the worth of any given mode of social life, the
next thing to notice is that they are both, in a sense to be explained, purely formal. Both,
Dewey says (DE, pp. 867),

point to democracy. The first signifies not only more numerous and more varied points
of shared common interest, but greater reliance upon the recognition of mutual interests
as a factor of social control. The second means not only freer interaction between social
groups (once isolated so far as intention could keep up separation) but change in social
habitits continuous readjustment through meeting the new situations produced by varied
intercourse.
The key expressions here are more variedshared common interest, freer interaction,
and changecontinuous readjustment. They all refer to formal characteristics. Nothing
is said about the substantial questions: the quality of the shared interests, the nature of the
results of the interaction, or the direction of the change. Pure formality is by no means
always a fault. Thus it is not, though it has often been said to be, a fault in the conception
of freedom, as freedom from, that it is negative; that it says nothing of what that freedom
is to be used to do. Nor is it a fault in the primary conception of democracy that it refers
only to procedures for taking or reversing decisions, not to the content or outcome of decisions taken. But it is at least odd, and to my mind also a fault, to offer such purely formal
characteristics as comprehensive criteria of social desirability: The problem is to extract
the desirable traits of forms of community lifeand employ them to criticize undesirable
features; and, From these two traits we derive our standard (DE, p. 83).
The nature of this oddity will be better appreciated if we compare Deweys position
with that of those for whom equality, as opposed to any kind of quality, is an independent
valueperhaps even the supreme value;12 for equality is, in the present sense, a formal
characteristic: it recruits a substantial associate only when we specify at what absolute
level and in respect of what the equality does or should obtain. Furthermore: if, and in so
far as, equality is accepted as an independent value, then, and to that extent, the way is open
for the sacrifices to the pursuit of equality of other and more substantial goods. The size of
the sacrifices actually required will, of course, depend: both on the weight given to equality
relative to those other values; and on what goods might in the particular situation in fact
be obtained. But suppose people are not willing to accept any sacrifices of other possible
goods for the sake of more equality; then, it follows necessarily, equality is not for them
an independent value.
Among the immediate consequences of the elementary analysis of the previous paragraph are that this commitment necessarily involves some degree of willingness: both to
hold back, or even to depress, the better off with no compensating advantage to the worse
off; and to hold back, or even to depress, everyone for the sake of more equality than might
otherwise be achieved.
Having thus emphasized unlovely implications of the single-minded pursuit of distributional equality, both charity and justice require me to notice at once that for many of
those who insist upon talking as if all inequality is self-evidently bad, such equality is
nevertheless not really an independent value. Concern for it is instead derived from other
value commitments which are for them more fundamental.13 Nor indeed is Dewey in the
end devoted to the mere numerousness and variety of interests as such. These indices serve
and can serve as criteria for the general excellence of a society only because, he persuades
himself, the more numerous and the more varied the interests, then, as a matter of fact, the
better (by other but here unspecified standards). He writes (DE, p. 83):

If we apply these considerations to, say, a criminal band, we find that the ties which hold its
members together are few in number. If we take, on the other hand, the kind of family life
which illustrates the standard, we find that there are material, intellectual, aesthetic interests
in which all participate.
Acceptance of the criteria proposed must also be facilitated by a certain carelessness as

to what they are supposed to be. The first question as originally proposed read: How
numerous and varied are the interests which are consciously shared? (DE, p. 83). Yet in
the Summary at the end of that same chapter we are told that the first of The two points
selected by which to measure the worth of a form of social life was the extent in which
the interests of a group are shared by all its members (DE, p. 99).
(c) The implications of these criteria

In examining these measures for the worth of any given mode of social life I first noticed
that, though Dewey is forever talking of his great society as democratic, this book contains
almost no references to any kind of political democracy.14 Nor does Dewey make much
attempt either to relate his ideals to any such established notion or to differentiate them
therefrom. Second, I pointed out that his indices are curiously formal. This formality would
perhaps be no fault if Dewey were attempting some account of democracy in an existing
sense. But it is very odd when what he is proposing is a comprehensive index of all the
excellences of an ideal society. My third concern is with the actual implications of the
criteria recommended.
For Dewey, we recall, the key questions are: How numerous and varied are the interests
which are consciously shared? How full and free is the interplay with other forms of association? (DE, p. 83). Since these are supposed to be criteria of democracy (DE, p. 87),
A democracy is more than a form of government, it is primarily a mode of associated living,
of conjoint communicated experience. The extension in space of the number of individuals
who participate is an interest so that each has to refer his own action to that of others, and
to consider the action of others to give point and direction to his own, is equivalent to the
breaking down of those barriers of class, race, and national territory which kept men from
perceiving the full import of their activity.
The second of those two sentences contains a characteristic emphasis upon what we have
to recognize as the third area of meaning. It also makes remarkable suggestions, that the
ideal society has an inherent drive to become a world community, and maybe that the bigger and more populous the community the better, too. The first sentence spells out the first
criterion. G.H.Bantock recently drew attention to the strong equalizing tendencies in the
direction of sameness in this asserted need for an increased community of experience
open to all.15 But Bantock missed the opposite tendency, towards diversification, which
Dewey discovers in his second criterion; for the passage just quoteda passage cited by
Bantock toocontinues (DE, p. 87):
These more numerous and more varied points of contact denote a greater diversity of stimuli to which an individual has to respond; they consequently put a premium on variation
in his action. They secure a liberation of powers which remain suppressed as long as the

incitations to action are partial as they must be in a group which in its exclusiveness shuts
out many interests.
But even if we do concede that this opposite tendency really is implicit in the original insistence upon maximum interplay with other forms of association, there is no getting away
from the truth of Bantocks contention that there are strong pressures towards equality of
outcome in the work of John Dewey;16 for if associations are good and democratic in so
far as their members share numerous and varied interests, and if education for democracy is
to be a matter of concentrating on the development of various but always shared interests,
then the variety of those shared interests, and the scope for independent individual development, necessarily must be limited correspondingly. It must, that is to say, be limited by
and to whatever happens to be the maximum attainable either by the least richly talented
or by the modal majority.
Maybe Dewey himself would have been unhappy about the full force of these implications. But he never comes to terms in this context with the truth that people vary enormously in all natural endowments. He is not so foolish as to deny the fact that we do. Yet
he somehow contrives not to notice its relevance to his first criterion of social merit. He
writes (DE, p. 90):
We cannot better Platos conviction that an individual is happy and society well organized
when each individual engages in those activities for which he has a natural equipment, nor
his conviction that it is the primary office of education to discover this equipment to its
possessor and train him for its effective use. But progress in knowledge has made us aware
of the superficiality of Platos lumping of individuals and their powers into a few sharply
marked off classes; it has taught us that original capacities are indefinitely numerous and
variable.
Fair enough. Our natural endowments vary in indefinitely many dimensions, and these
variations are continuous rather than separated by gulfs between sharply different categories. But these are no reasons for denying: either that the distances between the actual
extremes in any of these dimensions may beas they in fact very often areenormous;
or that there cannot beas again there in fact very often aresubstantial average differences in respect of some particular natural endowment as between one group of people and
another. Dewey does not draw either of these two negative conclusions, explicitly. They
do not follow. What he does do is simply to proceed as if he had. He recognizes no tension
between the tendency towards diversification, which he discovers in his second criterion,
and the drive towards an equality of outcome, which is built into the first; and he takes it
that the facts show that social organization means utilization of the specific and variable
qualities of individuals, not stratification by classes (DE, p. 91).
Allowas Dewey would certainly wish, and as must be truethat any defection from
ideal equality of opportunity wastes talent.17 Still a form of social utilization which makes
the fullest use of the specific and variable qualities of individuals does not thereby preclude stratification by classes; for even the sort of group which sociologists would be
prepared to admit as a social class can in its recruitment be wide open to the rising talents.
It can be open downwards too, to the lack of talents. To the extent that actual social classes
are in fact thus open, both upwards and downwards, and to the extent that any relevant

natural endowments are hereditable, the children of these open social classes are bound to
become as such members of a group of people distinguishable from other groups of people
by an average difference in respect of those particular endowments.18
Two or three paragraphs back I suggested, with charitable intent, that Dewey might
himself have been unhappy with the implications which Bantock and I both want to underline. No one, after all, should overlook the likelihood that he shared with us readers at least
some difficulties in determining exactly, or even not very exactly, what it was that he did
want to say. Yet he does himself later apply his criteria in ways that I do find disturbing.
For instance (DE, p. 122):
the idea of perfecting an inner personality is a sure sign of social divisions. What is called
inner is simply that which does not connect others, which is not capable of full and free
communication. What is termed spiritual culture has usually been futile, with something
rotten about it, just because it has been conceived as a thing which a man might have
internallyand therefore exclusively. What one is as a person is what one is as associated
with others, in a free give and take of intercourse.
Without those carefully chosen and worked examples which Dewey seems never to provide we are not, perhaps, entitled to feel sure precisely what this polemic against the private
is intended to condemn. There could be little educational point or profit here in a Wittgensteinian onslaught upon the (logically) private as such. It is, nevertheless, to be noted
that this interpretation could not be dismissed out of hand as a total anachronism, for in
discussing how words get understood, Dewey wrote: they acquire the same meaning with
the child which they have with the adult because they are used in a common experience of
both (DE, p. 15).
The obvious alternative interpretation is to construe Dewey as validly inferring from
the two proposed criteria of social excellence, and in particular from the first, the to my
mind appalling moral that everything (contingently) private is divisive, rotten, and intolerable. Remembering those rudimentary anticipations of the later Wittgenstein we might
see Dewey as reinforced in his determination to maintain this embargo by a rather murky
suspicion that what he deplores must in any case be somehow even logically vicious.
Yet does he really want to ban all private or solitary activities, all going apart from or
against the collective? I think here among other things of recent remarks by one of Deweys own most distinguished pupils: Possessing moral courage himself, Dewey took it for
granted that others had it; and The gabble in the academies about the vice of conformism
is empty and meaningless. What we must cherish is not agreement or disagreement but
intellectual independence, the courage to hold a position on the strength of the evidence no
matter what the baying of the crowds.19
(d) What is the appeal of these criteria?

Having, in the previous section, looked at implications of Deweys proposed criteria of
social excellence, it is now time to ask, in the fourth, how these could have seemed to him
so obviously right. The clue lies in a sentence which connects two passages already quoted
much earlier (DE, p. 83):

The problem is to extract the desirable traits of forms of community life which actually
exist, and employ them to criticize undesirable features and suggest improvement. Now in
every social group whatever, even in a gang of thieves, we find some interest held in common, and we find a certain amount of interaction and cooperative intercourse with other
groups. From these two traits we derive our standard.
From this it is easy to pick out the nerve of the argument. If and in as much as interests
held in common, and a certain amount of interaction and cooperative intercourse with other
groups, are defining characteristics of a society, then and for that reason the greater the
degree in which any society manifests these characteristics the more of a society it is, and
so the better it is as a society. It is not a sound argument, notwithstanding that arguments of
this same form seem to have persuaded many of the wise and good. Nor is it plausible to
suggest that the second of the two proposed characteristics is defining. The argument is not
sound, since you might as well say that, because a smoked salmon is by definition smoked,
it must be a better smoked salmon the more smoked it is. The suggestion is not plausible,
since there is no contradiction in speaking of a completely closed and isolated society, all
of whose members believewhat by the hypothesis I hereby stipulate to be in fact the
casethat it is the sole society there is.
Although the argument would be no better, the suggestion would be more plausible
if interaction and cooperative intercourse within the group itself were to be substituted
for interaction and cooperative intercourse with other groups. Indeed I cannot think of
any reason, apart from a general North American prejudice in favour of extraversion, why
Dewey should have introduced this reference toas it wereforeign affairs. An introverted version of Deweys second criterion might, however, be put forward as essential to
the idea of social relationship.
Back in chapter 1 Dewey himself explained the logical liaisons between the notions of
community, communication, and consensus (DE, pp. 45):
Men live in a community in virtue of the things which they have in common; and communication is the way in which they come to possess things in common. Individuals do not
compose a social group because they all work for a common end. The parts of a machine
workfor a common result, but they do not form a community. If, however, they were all
cognizant of the common end[,] and all interested in it[,] so that they regulated their specific
activity in view of it, then they would form a community. But this would involve communication. Consensus demands communication (commas supplied).
It is, I believe, both true and fundamental that both community and consensus presuppose
communicationproviding always that communication is not construed as necessitating a developed language. This is no place for attempting a full analysis of such notions.
It is sufficient and perhaps necessary simply to recall Humes example of two people
agreeing, without words and without contract, to row a boat together.20 The minimum of
communication required for such primal social acts is some mutual awareness, combined
with some mutual recognition of the possibility of reaction one to the other: There is, as
Dewey says, more than a verbal tie between the words common, community and communication (DE, p. 4).

But it is quite another thing, and much more disputatious, to say that a group of people
cannot constitute a community, and perhaps cannot even be in social relationships at all,
unless they have and are cognizant of a common end, are all interested in it, and all consciously regulate their activities with this common end in mind. To insist on this would,
surely, be an arbitrary exercise in high redefinition.21 It would also be vastly to overestimate
the actual role of such shared intentions in social living, for most people do not consciously
decide whether to become or to remain members of a society, as opposed to living as hermits. Hence they have little occasion to ask themselves what are the common ends of their
society, and how far they are themselves interested in these supposed collective purposes.
We cannot in such contexts too often remind ourselves of the importance of the unintended consequences of intended action. Dewey, I think, never went to school with Bernard
de Mandeville or with the great men of the Scottish Enlightenment. So let us ponder again
some famous words of Adam Smithwords so often drowned by the uncomprehending
ridicule of those whose socialism blinds them to an essential of social science:
As every individualendeavours as much as he can to employ his capitalthat its produce may be of the greatest value; every individual necessarily labours to render the annual
revenue of the society as great as he can. He generally, indeed, neither intends to promote
the public interest, nor knows how he is promoting ithe intends only his own gain, and
he is in this, as in many other cases, led by an invisible hand to promote an end which was
no part of his intention.
Furthermore, Smith continues, it is not22

always the worse for the society that it was no part of it. By pursuing his own interest he
frequently promotes that of the society more effectually than when he really intends to
promote it. I have never known much good done by those who affected to trade for the
public good.
3 Universal theory and particular stances

I have devoted most of this paper to Deweys notion of democracy, and its implications for
education. Since I am dealing with the book Democracy and Education, and since these are
the things which that book is supposed to be about, it would seem to be a natural and proper
distribution. But the consequence of this concentration is that I have been operating mainly
in what I cannot help but see as a disaster area. This raises for me a question which always
comes to mind whenever I attend to any of Deweys philosophical work. The question is,
brutally, how someone who seems to have been such a poor philosopher, and a poor writer,
can be so admired by former students of the quality of Ernest Nagel and Sidney Hook?
A similar question arises about the Australian philosopher, John Andersonfor those
of us who can judge him only by his published work have to rate this as much inferior to
that of such distinguished pupils as John Mackie, John Passmore, or David Armstrong. I
suggest that in both cases the explanation lies in the face-to-face impact of a personality,
and in a lifetime of commitments on innumerable public issues. Whether or not I am right
in my two scandalous judgments of philosophical calibre, and in my suggested answers to
the consequent questions, something does need to be said about some other recommenda-

tions which are, in Democracy and Education, presented, or misrepresented, as derivations
from a general theory of education for democracy. Many of these must have had, and have,
much wider appeal than that theory; although the appeal of some is not, of course, to the
same people as the appeal of others.
(a) The unintended in education

It is, for instance, remarkable that someone so inclined when traversing the higher reaches
of theory to exaggerate the importance of intention should, reverting to concrete educational situations, maintain a wholly salutary emphasis upon unintended learning and
often unintended teaching: this unconscious influence of the environment is so subtle
and pervasive that it affects every fibre of character and mind; and Adults are naturally
most conscious of directing the conduct of others when they are immediately aiming so
to do. But the more permanent and influential modes of control are those which operate
from moment to moment continuously without such deliberate intention on our part (DE,
pp.18, 26).23
(b) Confusion about the social

Notice next, and this has more connecting with his official theme, that Dewey rejects the
suggestion that an individuals tendencies are naturally purely individualistic or egoistic,
and thus anti-social. He points out that we are all also interested in entering into the
activities of others and taking part in conjoint and cooperative doings (DE, pp. 23, 26).
Unfortunately he makes no distinction between: what is anti-social, in the sense of damaging to others; and what is anti-social, in the sense of opposed to all essentially social institutions. He is also inclined to mistake it that whatever is in that fundamental sense essentially
social must be collectively directed and controlled. Dewey thus falls into two nowadays
extremely popular misconceptions: that the private is necessarily antisocial, in the sense of
damaging to others; and that the social, in either sense, is necessarily collectivist.
For instance, meditating applications of electrical science to means of communication, transportation, lighting of cities and houses, and more economical production of
goods, he comments (DE, p. 201; italics original):
These are social endsand if they are too closely associated with notions of private profit,
it is becausethey have been deflected to private uses: a fact which puts upon the school
the responsibility of restoring their connection with public scientific and social interests.
Wait a minute: the fact that certain essentially social arrangements make private profits
possible has no tendency to show that these arrangements are in the ordinary and superficial sense anti-social, or that they do not at the same time fulfil all sorts of other worthy
public and private purposes; and, furthermore, I want to say: not only are Deweys implicit
arguments for socializing conclusions unsound as arguments; but there is also in the public
interest in fact a great deal to be said for private profitespecially perhaps when opposed
to todays most favoured actual alternative, public loss.
(c) Collectivism and centralization

Deweys collectivist inclinations were not at this time specifically statist.24 Certainly he
did always support the American public schoolsthis magnificent institutionagainst
all private alternatives: he speaks with enthusiasm of that growing and finally successful
warfare against all the influences, social and sectarian, which would prevent or mitigate
the sway of public influence over private ecclesiastical and class interests.25 Certainly he
moves much too quickly from the need for the support of the state to a movement for
publicly conducted and administered schools. The unreconstructed liberal will wish that
Dewey had attended to the warnings of John Stuart Mill: One thing must be strenuously
insisted upon; that the government must claim no monopoly for its education either in the
lower or the higher branches,26
But we should also remember: that the system which Dewey favoured so strongly is in
fact remarkably decentralized; that it appears to have room for a lot of selection on academic grounds; and that one of its units has recently become the first educational authority
in the world to introduce vouchers as an instrument for extending effective family choice.
All these things must give pause to anyone hoping to recruit Dewey as a posthumous
patron of the enrags of Education for Democracy.
(d) Philosophical insights and the teaching situation

Much of Deweys appeal to those who really care about teaching surely lies in such excellent down-to-earth recommendations as: There must be more actual material, more stuff,
more appliances, and more opportunities for doing things. (DE, p. 156: italics original).
This particular recommendation is presented as the outcome of philosophizing independent
of the main official theme of the book. The cruces are the priority of knowing-how over
knowing-that, and the wrongness of any receptive blank-paper account of the acquisition
of knowledge: The knowledge which comes first to persons, and that remains most deeply
ingrained, is knowledge of how to do (DE, p. 184: italics original); while It would seem
as if five minutes unprejudiced observation of the way an infant gains knowledge would
have sufficed to overthrow the notion that he is engaged in receiving impressions (DE,
p. 271).
(e) Dewey and Whitmans democracy

Another excellent recommendation is that it is easier to make economic than political history come alive for children. But here we return to the main theme: Economic history is
more human, more democratic, and hence more liberalizing than political history (DE,
pp. 21516: italics supplied). It is my cue for a final suggestion. The democracy of which
Dewey talked, Walt Whitman sang:
Yet utter the word Democratic, the word En-Masse.27
I speak the password primeval, I give the sign of democracy,
By God! I will accept nothing which all cannot have their counterpart of on the same
terms.28

It is in Democracy(the purport and aim of all the past)
It is in the life of one man or one woman todaythe average man of today.29
I heard that you asked for something to prove this puzzle to the new world,
And to define America, her athletic Democracy,
Therefore I send you my poems that you behold in them what you wanted.30
Notes
1 Democracy and Education by John Dewey, New York, Macmillan, 1916. All references to DE
below will be to the 1964 third printing of the 1961 Macmillan Paperbacks edition.
2 Quoted, with no reference given, by John L.Childs, The educational philosophy of John Dewey,
in P.A.Schiipp (ed.), The Philosophy of John Dewey, New York, Tudor, 2nd edn, 1951, p. 417. The
phrase is not to be found in any contribution by Dewey to the Library of Living Philosophers, for
in this case there is no Autobiography, but only a Biography compiled with his help by his daughters.
3 David Rubinstein and Colin Stoneman (eds), Education for Demo-cracy, Harmondsworth and
Baltimore, Penguin Education, 2nd edn, 1972. All references to ED, below, will be to this edition.
4 See his Sense and Sensibilia, Oxford, Clarendon, 1962, p. 127. The whole sentence reads: There
are, however, a few notoriously useless wordsdemocracy, for instanceuses of which are
always liable to leave us in real doubt what is meant; and here it seems reasonable enough to say
that the word is vague (italics original).
5 Who first said this, and where?
6 Reported in East Europe for July 1957, p. 56. I borrow this reference from Sidney Hook, Political
Power and Personal Freedom, New York, Collier, 1962, p. 147.
7 I collected this one myself from the Dar-es-Salaam press, the following day, 8 July 1967. Although
Dar is scarcely one of Ian Flemings exciting cities I have no difficulty at all in understanding
that and why those who have settled there even into penury not only do not want to be pushed
about for their own advantage but also hate the thought of returning to, by comparison, the
excruciating boredom of a Tanzanian village.
8 See on this, in the French Revolution of 1789, J.L.Talmon, The Origins of Totalitarian Democracy, London, Secker & Warburg, 1952. When Lenin first developed his ideas for a party of a new
type he himself contrasted what has since come to be called democratic centralism with genuine
democratic control; which last phrase he then interpreted in our first sense. See his What Is To Be
Done?, translated by S.V.Utechin and Patricia Utechin, London, Panther, 1970, pp. 1634, 1856;
and compare my Russell on Bolshevism, in a forthcoming collection edited by George W. Roberts, to be published by Allen & Unwin.
9 A.de Tocqueville, Democracy in America, edited by Richard D. Heffner, New York, New American Library, 1956, part I, ch. III.
10 There is even one, Donald McIntyrefrom the Labour Party of Ellen Wilkinson rather than that
of Edward Shortwho dares to argue that assessmentwhether by written examinations or by
other techniquesis an integral part of effective teaching (ED, p. 164). He is still too modest,
for there is, surely, a logically necessary connection between assessment of some sort and both
intentional teaching and intentional learning. See Teaching and testing, in my Sociology, Equality and Education, London, Macmillan, 1976.
11 J.Dewey and E.Dewey, Schools of Tomorrow, New York and London, Dutton and Dent, 1915, pp.
303ff.
12 On the back cover of ED, after essential reading The Times Educational Supplement, and
before an enthusiastic puff from Clyde Chitty (Morning Star), we read words from John Vaizeys

review in the Listener: the egalitarian passion of these writers is a passion that relies in the
main on hard, cold, realistic scholarship.
Certainly there is egalitarian passion in plenty. But there is also a great deal which equally
certainly is not hard, cold, realistic scholarship. For example: the two editors, after asserting that,
Throughout history the middle and upper classeshave given to the working classes as little and
as poor an education as possible, in support deploy only two quotations. One of the quotations
does nothing to vindicate their denunciations. The other maintains, in one particular case, its flat
contrary (p. 7). Again, Anthony Arblaster quotes a famous Austin paper, and then denounces Austin for failing to take points which, had Arblaster been willing to read only three more pages, he
would have found there made by Austin himself (p. 36). And so on.
For further details of these and other manifestations of the ideals and illusions of Penguin
Education, consult the work mentioned in note 10 above.
13 Some, for instance, being at bottom utilitarians, believe only because they judge that the marginal
utility of additions is less the more anyone has already. See A.M.Quinton, Utilitarian Ethics,
London, Macmillan, 1973, pp. 756; and compare C.Jencks, Inequality, London, Allen Lane,
1973, pp. 910.
14 One, at page 87, begins: The devotion of democracy to education is a familiar facta government resting on popular suffrage cannot be successful unless those who elect and who obey their
governors are educated. But this rationale is dismissed as superficial: there is a deeper explanation. A democracy is more than a form of government; it is primarily a mode of associated
living, of conjoint communicated experience. So Dewey proceeds, in terms quoted earlier in the
text, to apply his official criteria, and to promise the breaking down of those barriers of class,
race, and nationality which kept men from perceiving the full import of their activity.
The other political references are equally uninterested. At page 260 Dewey writes: In what
is termed politics, democratic social organization makes provision for this direct participation in
control. If this raises the spirits of participatory democrats, they will certainly be both surprised
and disappointed to learn that the sentence immediately preceding suggests that the free or voluntary activity of control can, apparently, mean as little as peoples inert concern about the ends
that control their activity; for that sentence reads: In the degree in which men have an active
concern in the ends that control their activity, their activity becomes free or voluntary and loses
its externally enforced and servile quality, even though the physical aspect of behaviour remain
the same (italics supplied).
It is a relief after this to read, on page 301, at the end of a paragraph on German philosophy:
Political democracy, with its belief in the right of individual desire and purpose to take part in
readapting even the fundamental constitution of society, was foreign to it. Again, civic efficiency is defined, on page 120, as involving ability to judge men and measures wisely and to take
a determining part in making as well as obeying laws.
15 Equality and education, in B.Wilson (ed.), Education, Equality and Society, London, Allen &
Unwin, 1975, pp. 11920.
16 Ibid., p. 119: italics original. Jencks, who is very clear that and why equality of opportunity cannot produce equality of outcome (op. cit., p. 37), has a whole chapter on Inequality in cognitive
skills. It ends with a list of ways in which cognitive inequality could or could not be reduced.
Recognizing that this possible ideal might be achieved by arranging that the ablest children should
have the least schooling and the dullest the most, Jencks himself stands up for what he calls equal
opportunityeccentrically construed as implying that everyone should get as much schooling
as he wants (ibid., p. 107: italics supplied). But, be warned now, if such cognitive inequality
were a principal cause of inequality in other realms, this traditional doctrine might need reexamination (ibid., pp. 10910).
17 Dewey does often think of scarce talents as a resource too precious to be left unexploited. For
instance, in a chapter on Vocational aspects of education, he says: since slaves were confined

to certain prescribed callings, much talent must have remained unavailable to the community, and
hence there was a dead loss (DE, pp. 3089: and compare pp. 94, 11819).
18 This conclusion is, surely, likely to be of practical relevance in a society such as ours which has
had considerable social mobility for some time. Certainly we have no business simply to assume
that the children either of every group of people or of every recognized social class must be on
average in every respect as well or ill endowed as the children of every other group or class. Yet
the Penguin Educationalist, Peter Mauger, like so many others, takes it that, if the children of one
social class are below average in their performance under some system of selection for talent,
then that fact is by itself sufficient to show that the system is inefficient, unfair, and ought to be
abolished. It does not follow.
He begins: Research has shown beyond a shadow of a doubt that working-class children are
grossly under-represented in grammar schools; and he then gives a reference to work by J.W.B.
Douglas. Mauger now proceeds immediately to his conclusion: These figures alone are strong
enough arguments for the abolition of maintained grammar schools and direct grant schools. If,
after all these years of experiment in selection procedures, they cant do better than this (ED,
pp. 1301).
19 Sidney Hook, Education and the Taming of Power, London, Alcove, 1974. The quality which
Dewey took for granted is one which from personal knowledge I can say that Hook too possesses
in abundance.
20 Treatise, III (ii) 2; page 490 in the Selby-Bigge edition, Oxford University Press, 1906.
21 The useful expressions, high redefinition and low redefinition, for those which, respectively,
increase and reduce the required qualifications, were introduced by Paul Edwards in Bertrand
Russells doubts about induction, first published in Mind, 1949, and reprinted in Antony Flew
(ed.), Logic and Language, Oxford, Blaekwell, First Series, 1951.
22 The Wealth of Nations, IV (ii), p. 400, vol. 1, Everyman edition. Compare F.A.Hayek, The results
of human action but not of human design, in his Studies in Philosophy, Politics and Economics,
London, Routledge & Kegan Paul, 1967.
23 Dewey earlier contrasts indirect education, by direct sharing in the pursuits of grown-ups, with
formal education, in Intentional agenciesschools; warning that the latter easily becomes
remote and dead (DE, p. 8). I recall here a tale told by a former Ghanaian colleague. He was
himself present at a Convention Peoples Party demonstration against the University of Ghana, at
which the slogan was raised: Down with bookism!
24 They became more so later. Hook claims to have played a part in this change: I had contended, and
still do, that only in a democratic socialist economy in which housing, schooling and vocational
opportunities are planned in relation to the needs of individuals, and not made dependent upon
the vagaries of the market, could the educated ideals of Democracy and Education be realised.
By the time he published his Liberalism and Social Action Dewey had come round to accepting
this (op. cit., p. 175).
It is clear that Hook was working on a subject already more than half converted, disinclined
to criticize socialist ideas. For instance: Dewey asserts that employees cannot have insight into
the social aims of their pursuits unless the industries in which they work are under collective
control. At present, he says, The results actually achieved are not the ends of their actions, but
only of their employers (DE, p. 260: italics original). Yet why on earth should anyone accept this
contemptuous and doctrinaire contention, that people working in individually owned bakeries,
plastics factories, or whatever else, cannot appreciate that the products of their work meet their
own and other peoples needs for these products?
25 J.Ratner (ed.), Democracy in Education, London, Allen & Unwin, 1941, p. 63.
26 Principles of Political Economy, V (xi). Those who have heard that Mill was a Fabian socialist
before their time should study this whole chapter. It contains uninhibited Selsdon Group attacks
on all monopolies, but especially state monopolies, as well as on excesses and abuses of trades

union power. Compare my J.S.Mill: socialist or libertarian?, in M.Ivens (ed.), Prophets of Freedom and Enterprise, London, Kogan Page, 1975.
27 Ones self I sing, in E.de Selincourt (ed.), Selected Poems, Oxford University Press, 1920, p. 1.
28 Song of myself , ibid., p. 44.
29 I was looking a long while, ibid., p. 297.
30 To foreign lands, ibid., p. 3.
6
John Deweys philosophy of education
R.S.Peters
Introduction
It would be tempting to regard John Deweys philosophy of education as an extrapolation
of key features of learning situations in the old rural life, in which he was nurtured, to
schooling in the industrial society which developed during his lifetime. Dewey experienced the new schooling firsthand as a not very successful teacher and was appalled by the
rote-learning, regimentation, and irrelevance to life that characterized so much that went
on. His philosophy, it might be said, was an attempt to introduce into this new institution
the problem-solving, do-it-yourself method of the learning of his boyhood, together with
the close link between learning and living and the sense of contributing to a social whole
permeated by shared experiences.
No doubt there is some truth in this suggestion; for peoples views about an educational situation are very much influenced by early models which they encounter. Michael
Oakeshott, for instance, confessed at the end of an article on Learning and teaching1 that
he owed his recognition of the values of patience, accuracy, economy,elegance,and style
to a sergeant gymnastics instructor, not on account of anything he said, but because he
exemplified them.
The key to understanding Deweys philosophy of education, however, is not just his
early experience nor the obvious point that he was a pragmatist who applied the doctrines
of Charles Peirce and William James in a straightforward way to education. Rather it is
the realization that he was, for a long time, a Hegelian who later became converted to
pragmatism. Dewey, like Hegel, could not tolerate dualisms. He had a passion for unifying doctrines that, on the surface, seemed irreconcilable. Pragmatism, and especially its
emphasis on scientific method, together with categories of thought extrapolated from biology, seemed to him the key to unification. It also seemed a natural extension of his early
experiences of problem-solving.
In his educational theory this passion for unification, for getting rid of dualisms, had
ample scope as the titles of his books indicate: The Child and the Curriculum,2 The School
and Society,3 Interest and Effort in Education,4 Experience and Education5 and so on. This
quest for unity, which Dewey substituted for the despised quest for certainty, explains why
Dewey was not a wholehearted supporter of the progressive movement in America and
ended up by writing his Experience and Education, which was highly critical of some of
its practices.
1 Individual growth and shared experiences

The exception might seem to be Democracy and Education,6 which is a puzzling book, for
there is plenty about education in it but very little about democracyno proper discussion
of liberty, equality and the rule of law, no probing of the problems of representation, participation, and the control of the executive. The explanation of this is that Dewey viewed
democracy mainly as a way of life; he was not particularly interested in the institutional
arrangements necessary to support it. This way of life, he claimed, had two main features.
First, it was characterized by numerous and varied shared interests and concerns. These
play an important role in social control. Second, there is full and free interaction between
social groups, with plenty of scope for communication.7 This is surely a strange characterization of democracy. What is significant about it, however, is the emphasis on the social.
Dewey later says:8
And the idea of perfecting an inner personality is a sure sign of social divisions. What is
called inner is simply that which does not connect with otherswhich is not capable of
free and full communication. What is termed spiritual culture has usually been futile, with
something rotten about it, just because it has been conceived as a thing which a man might
have internallyand therefore exclusively. What one is as a person is what one is as
associated with others, in a free give and take of intercourse.
There is nothing particularly surprising about this view of Deweys background in idealism. Indeed it is very reminiscent of Bradleys attack on individualism in his famous chapter on My station and its duties.9 Also the distinction between the public and the private
has never been so sharply drawn in American society as in Europe. A modern symptom
of this is that in a typical American township they have no walls to their gardens. But the
puzzle is to reconcile this emphasis on shared experiences and social approval and disapproval, as one of the main moulders of character, which Dewey emphasizes, with his view
of education as individual growth.
Much has been written about the unsatisfactoriness of this biological metaphor which
Dewey used to impose unity on his theorizing. He argued that growth does not have an end
but is an end. Thus education is not necessarily a matter of age; for education means the
enterprise of supplying the conditions which ensure growth, or adequacy of life, irrespective of age. Living has its own intrinsic quality, whether in youth or in maturity, and the
business of education is with that quality.
But, it is usually objected, how can growth provide criteria of this quality? Did not
Napoleon or the Marquis de Sade grow? Dewey faced this problem in Experience and
Education10 and argued that growth in efficiency as a burglar, as a gangster, or as a corrupt
politician does not lead to further growth. Growth in general is retarded by such limited
forms of growth. This answer did not help him much as many have pointed out, with
imaginative sketches of the developing life-styles of burglars, train robbers, etc. What he
needed was other criteria by reference to which desirable and undesirable forms of growth
could be distinguished.
In actual fact I think that Dewey did have other criteria of value for education. His metaphor of growth, like his other concepts of interaction and continuity, are just part of
his conceptual apparatus, taken from biology, which symbolize his insistence that man is
John Deweys philosophy of education 67

part of the natural world. He could not tolerate the dualism, found in thinkers such as Kant
and Descartes, between man and nature any more than he could tolerate other dualisms.
But later in Democracy and Education11 he reached what he called a technical definition of education as that reconstruction or re-organization of experiences which adds to
the meaning of experience, and which increases ability to direct the course of subsequent
experience. He went on to say that an activity which brings education or instruction with
it makes one aware of some of the connections which had been imperceptible. He stressed
the addition in knowledge both of the child who reaches for a bright light and gets burned
and of the scientist learning more about fire in his laboratory. In brief, Deweys main concern was with growth in practical knowledge, in the development of critical intelligence as
described in his earlier popular book How We Think.12
But Dewey, as a pragmatist, was also interested in what he called the other side of an
educative experience which is the added power of direction or control. He contrasted
this with aimless activity, on the one hand, and he singled out acting under external
direction as a classic example of this, and routine activity on the other hand, which
only increases skill in doing a particular thing. This could be interpreted as a way of
stressing the virtue of autonomy, of self-initiated action which is the outcome of independent thought. Indeed, Dr Dearden, in his The Philosophy of Primary Education,13
sees autonomy as the ethical value which is embedded in the whole growth ideology.
What is common to both is the notion of self-originated activity. This is certainly a
valid interpretation of many other more individualistic theorists who made use of the
metaphor of growth. But I doubt whether it completely fits Dewey with his continued
emphasis on shared experiences and communication. I think his ideal was much more
that of a group of dedicated, problem-solving scientists, who were united by their shared
concerns and willingness to communicate their findings to each other. The lonely will
of the individual was anathema to him,14 though in chapter 5 of Experience and Education he does extol the virtue of freedom understood as self-control whilst pointing out
that too much freedom understood negatively as the absence of social constraints may be
destructive of shared co-operative activities. I think that he more or less took for granted
the value of individual self-determination but was more concerned to stress the values of
co-operative problem-solving as an antidote to the extremes of individualism in the old
pioneer period.
This interpretation of Deweys concept of education not only gives more determinateness to his somewhat nebulous metaphor of growth; it also explains the link which
he forged between education and his rather strange conception of the democratic way of
life. His emphasis, in the latter, on numerous shared interests and communication makes
sense if it is seen as a projection of features of the kinds of communities in which he
worked and lived, for Dewey was an academic who, as well as writing forty books and
about 700 articles, was constantly founding and joining groups concerned with various
forms of social and educational reform. Thus growth for him was not growth in any
direction which would be consistent with his claim that desirable growth is that which
permits more growth; it is rather growth in practical critical thought, which opens up
the possibility of more control of the environment. But this is not something which the
individual does on his own. In his early days at Michigan one of Deweys colleagues
was G.H.Mead, whose theories about the social nature of the self and the social deter-

minants of thought influenced Dewey profoundly. These theories strengthened Deweys
Hegelian convictions about the social nature of man and supplied support for his own
distinctive brand of pragmatism, with its emphasis on shared experiences. It enabled
him to argue that growth, properly understood, can only flourish in a democratic environment. Indeed, to use a Platonic metaphor, for Dewey the democratic way of life is
growth writ large. There is unity discernible beneath the appearances of democracy
and education.
2 Individual interest and external direction

Deweys attempt to transcend dualisms is nowhere more apparent than in his treatment of
the teaching situation.
(i) Aims of education

There is first of all his treatment of aims of education in which he attacked the false
dichotomy between means and ends which he exposed at greater length in his Human
Nature and Conduct.15 But he had additional concerns in his educational writings. First, he
insisted on the intrinsic value of educational activities. They are not merely unavoidable
means to something else.16 Second, he maintained that good aims arise from what is going
on, from the purposes of the pupil. They must not be externally imposed, or ready made.
Nevertheless he did not advocate a kind of free for all in which any aims are accepted if
they arise in this way. They must be capable of translation into a method of co-operating
with those undergoing instruction; they must lend themselves to the construction of specific procedures. And who is to be the judge of this unless it is the teacher? Also Dewey
realized that such aims do not spring spontaneously from the nature of the child. Indeed,
he criticized Rousseau for making Nature his God. They are moulded by what he called
the social medium. Although he was critical of imitation as an important factor in the
social medium he admitted the large influence of social approval and disapproval.17 Thus
all along the line Dewey tried to combine the progressive child-centred approach with
what he had learnt from Mead and with what was in his bones as a Hegelian. He resisted
external direction and imposition but insisted on the importance of external approval and
encouragement. He thus achieved some kind of reconciliation between the progressive and
traditional views of teaching.
(ii) Teaching methods

This attempt to get rid of dualisms was made even more explicit in his Experience and
Education.18 He was at pains to point out that he was not suggesting a passive or spectatorial role for the teacher. Indeed, he argued that basing education upon personal experience may mean more multiplied and more intimate contacts between the mature and the
immature than ever existed in the traditional school, and consequently more rather than
less guidance by others.19 In their account of Deweys laboratory school Mayhew and
Edwards insist that Those planning the activities must see each child as an ever changing
person. They must carefully select and grade the materials used, altering such selection,

as is necessary in all experimentation,20 (This was practicable because of the extremely
favourable teacher-pupil ratio. The school started with 3 regular teachers for 32 children;
rose to 16 teachers for 60 children, and ended with 23 teachers plus 10 assistants for 140
children!)
Dewey himself described this careful grading and selection of material in terms of
his two criteria of educative experiences, interaction and continuity. He used the term
interaction, rather than more homely terms such as needs and purposes of the child,
not purely because of his desire to create some kind of biological unity between the processes of education and those of life but also because too many progressives, in his opinion,
had neglected the objective conditions of situations and the role of the teacher in arranging
for them to match the internal conditions of the child. Similarly continuity was stressed
because it was not sufficient for the child to be interested in anything; interests had to
be explored which were rich in possibilities for future experiences. So guidance by the
teacher was substituted for the external direction of traditional methods, and because
interests arose from the child, deriving from his impulses of investigation and experimentation, constructiveness, expressiveness and the social impulse,21 the approach could claim
to be child-centred.
The method of learning which conformed to these criteria of educative experiences
was that of problem-solving, a detailed account of which was given by Dewey in How We
Think. This stress on problem-solving as a method was later taken up by Kilpatrick and
formalized in the project method. Dewey was favourably disposed towards it but did not
become a passionate advocate of it. To be fair to him he was always very guarded about
details of teaching methods. He confined himself to generalities, knowing that details of
implementation must vary with individuals.
(iii) Social control and the role of the teacher

Deweys account of the social control of the teacher exhibited the same tendency towards
unification. He tried to transcend the dichotomy between the keeping order view of the
traditional school and the self-imposed discipline advocated by the progressives. He compared children in a classroom to their participation in a game. Games involve rules and
children do not feel that they are submitting to external imposition in obeying them. The
control of the actions of the participating individuals is affected by the whole situation in
which individuals are involved, in which they share and of which they are co-operative or
interacting parts.
The teacher exercises authority in such a situation as the representative and agent of the
interests of the group as a whole. If he or she has to take firm action it is done on behalf of
the interests of the group, not as an exhibition of personal power. In the traditional school
the teacher had to keep order because order was in the teachers keeping instead of residing in the shared work being done. In the new schools the main job of the teacher is to
think and plan ahead so that knowledge of individuals may be married with knowledge
of subject-matter that will enable activities to be selected which lend themselves to social
organization. Thus the teacher loses the position of external boss or dictator but takes on
that of leader of group activities.22
3 The content of education and the role of the school

Dewey is sometimes classified with those progressives who have extolled following the
interests of the child at the expense of subject-matter. This is completely to misunderstand
his position, for he was too much of a Hegelian to ignore the importance of a societys
cultural heritage which he described as the ripe fruitage of experience. But, again as
a Hegelian, he strove to remove the dichotomy between both the child and the curriculuim, and the school and society. On the one hand, therefore, he insisted that the curriculum should embody what he called the sociological and the psychological principles. The
sociological principle demanded that the pupil be initiated into the customs, habits, values,
and knowledge which constitute the culture of a community. The psychological principle
demanded that this should be done with due regard to the pupils individual needs, interests
and problems.
On the other hand he believed passionately that the curriculum should be socially relevant. It should contribute to making children active members of a democratic society.
Indeed, on this theme Dewey waxed almost mystical and poetic:23
When the school introduces and trains each child of society into membership within such
a little community, saturating him with the spirit of service, and providing him with the
instruments of effective self-direction, we shall have the deepest and best guarantee of a
larger society, which is worthy, lovely and harmonious.
Let us, therefore, consider in more detail his resolution of the dichotomies between the
child and the curriculum and between the school and society.
(i) The child and the curriculum

Most of what Dewey wrote about the curriculum related to the elementary school and
much of it seems rather dated. But it illustrates well his approach. He stressed, first of all,
the importance of practical activities such as sewing, cooking, weaving, carpentry and
metalwork. These conformed to the sociological principle because they were basic to life,
being concerned with food, clothing, etc., and thus part of the cultural heritage. They also
conformed to the psychological principle for two reasons. First, Dewey was convinced
that children are interested in them. Second, they embody motor activities which Dewey
considered to be closely connected with mental development as a whole. Also, from an
educational point of view, they were capable of providing continuity in that they could
open up all sorts of other fruitful studies. As he put it: You can concentrate the history of
all mankind into the evolution of the flax, cotton and wool fibres into clothing.24
In addition to practical activities he included some traditional subjects in the curriculum with the proviso that they should be related to his concept of man as a problem-solving
animal concerned with control over his environment. Thus he regarded geography as being
of particular importancebut as a way of gaining in power to perceive the spatial, the
natural connections of an ordinary act. History was acceptable, too, as a way of recognizing the human connections of ordinary acts.25 And both, of course, must start from the
childs immediate interestsgeography must move outwards from local geography and

history from some present situation with its problems.26 Science is, of course, included,
but subject to the same sort of provisos. It should be taught with the psychological principle
in mind and start from the everyday experience of the learner. There was too much of a tendency to teach it in the logical order of the developed study. Above all, science should be
taught as the agency of progress in action, for it opens up new ends as well as helping mankind to achieve existing ones. Because of science man can now face the future with a firm
belief that intelligence properly used can do away with evils once thought inevitable.27
Finally the curriculum should include communication skills such as reading, writing,
mathematics, and foreign languages. These appealed to the childs impulses to express
himself and to share his experiences with others. So the best time to teach him the techniques of communication is when the need to communicate is vitally important to him.
These communication skills should be taught incidentally as the need arose.
(ii) The school and society

There were two aspects of Deweys attempt to resolve the dualism between the school and
society. The first dealt with the relationship of the school to the home and surrounding community, the second with its relationship to the wider society which the pupil would enter on
leaving school. On the first aspect, as I said at the beginning, Dewey was greatly impressed
by the informal type of learning that went on in the home and in the small rural communities that were passing. He frequently contrasted this natural way of learning, in which there
was no separation between learning and life, with the artificial drills and recitations of
formal schooling. His plea was that there should be an indissoluble link between learning
in school and learning out of school.28 Deweys insistence that the school itself should be
a real community, exhibiting numerous shared interests and open communication, was his
answer to the other question of the schools relationship to the wider society. The school
itself should be a miniature democracy, according to his understanding of democracy. He
saw this type of school not only as valuable in itself, because of the quality of life that it
made possible, but also as the springboard to social progress. Dewey took a prominent part
in the current controversy about Trade Schools and vocational education.29 He deprecated,
of course, the split between the practical and the liberal which reflected an undesirable
type of class-structure. He objected to the implicit suggestion that education should be
made subservient to the demands of interested manufacturers. Nevertheless his solution
was typically one in which the dualism between vocational and liberal education could be
resolved; for he argued that if more practical activities were introduced into schools, education would be through occupations and not for occupations. He advocated the introduction
of processes involved in industrial life to make school life more active, more impregnated
with science, and more in touch with the world. This should be part of everybodys education, not just a special provision for those who were singled out to become the modern
equivalents of hewers of wood and drawers of water. Above all, a different attitude to work
should be developed so that young people would become imbued with a sense of community service instead of working only for private gain. It should train power of readaptation
to changing conditions so that future workers would not become blindly subject to a fate
imposed upon them.30

4 General comments
What is to be made of this intellectual edifice? For opinions about it are very varied. I once
tried to get an eminent American philosopher interested in the philosophy of education. He
grunted and remarked that John Dewey had set that subject upand killed it stone dead!
On the other hand Sidney Hook, another eminent American philosopher, published a book
as recently as 1973, in which he included several essays in defence of Dewey.31 He saw
Deweys philosophy as providing a middle road between radicals such as Reimer, Goodman and Illich, and the post-sputnik traditionalists such as Vice-Admiral Rickover. I agree
with Sidney Hook that such a middle road is necessary, but I do not find the one signposted
by Dewey particularly convincing or congenial; for his way of resolving various dualisms
by his account of the growth of the problem-solving man has the character of a panacea
which involves both distortion in the sphere of what he called the sociological principle
and a romantic idealization in the sphere of the psychological principle. The dualisms are
not in fact resolved. Let me explain these criticisms before addressing myself to an estimate of his unifying ideal.
(i) The sociological principle

Dewey admitted the importance of making the child aware of his cultural heritage but only
on the condition that he should be introduced to it in a way which stressed its relevance to
present practical and social problems. This is understandable as said against unimaginative
rote-learning of classical textbooks, but, if taken seriously, is a good recipe for failing to
understand what we have inherited, for it fails to take account of the degree of autonomy
which some traditions of inquiry have from contemporary practical problems. Understanding depends upon entering imaginatively into the mind of those who have contributed to
these traditions and grasping what their problems were as arising from them. Copernicus
and Kepler, for instance, were both working within the Pythagorean tradition. The heliocentric theory emerged because it was mathematically simpler; Keplers second law of
planetary motion was lighted on in the course of speculations about the music of the heavenly bodies. To stress the relevance of these momentous advances for navigation or space
travel does nothing towards understanding them as theories. It may be said that the point
is not so much to understand such theories thoroughly, but to use them. But a failure to
understand properly the problems with which people in the past have been concerned often
leads to absurdities in attempts to use them. Piaget, for instance, was greatly influenced
by Kant. Piagets theories are widely applied in the educational sphere in a ham-handed
way because the educationalist in question has no conception of what Kant was about in
his critiques. The school, surely, should not concern itself only with what is relevant to
contemporary problems. It should also distance itself a bit from these and introduce children to speculations about the world in science, and to insights into the human condition in
literature and history, which are of perennial significance. The dualism is there and gives
rise to continuing tensions.
Deweys view of the teacher, who is societys agent for the transmission and development of its cultural heritage, is also unsatisfactory, for it slurs over the dualism between the
teachers position as an authority and the legitimate demand for participation. A teacher

is not just a leader in a game, like a football captain. In a game most of the participants
know how to play; but pupils come to a teacher because they are ignorant, and he or she
is meant to be, to some extent, an authority on some aspect of the culture. This disparity
between teacher and taughtespecially in the primary schoolmakes talk of democracy
in education problematic, unless democracy is watered down to mean just multiplying
shared experiences and openness of communication, as by Dewey. If democracy is to
include, as it usually does, some suggestion of participation in decision-making, we are
then confronted with current tensions underlying the question of how much participation
is compatible with the freedom and authority of the teacher.
Dewey himself never paid much attention to institutional issues. This was not just
because he lived before the days when participation became an issue. It was also because
his attitude towards the democratic way of life was semi-mystical. When the emotional
force, the mystical force, one might say, of the miracles of the shared life and shared experience is spontaneously felt, the hardness and concreteness of contem-porary life will be
bathed in a light that never was on land or sea.32 I wonder if he always felt like this about
sitting on committees!
(ii) The psychological principle

Deweys treatment of the psychological principle was equally unsatisfactory; for it
combined a conception of the child, which was almost as idealistic as his conception of
democracy, with a too limited view of what he called the social medium. This led him
to oversimplify the dualism between what he called internal conditions and what is the
result of social influences. Dewey was impressed, as I have reiterated, by the informal
learning that went on in the home and in the local community and wanted to forge a link
between this sort of learning and learning at school. But he did not ask the questions which
home? and which local community?, for sociologists have catalogued the vast disparities
that exist between homes in this respect.
Deweys account of the ideal educational situation assumed, to start with, an impulse
to investigate and experiment, as well as a social impulse from which co-operation stems.
Maybe most of the children in his Laboratory School had such impulses. Maybe all children have them at birth. But by the time they get to school it is noticeable how many
children seem to lack these impulsesand this is probably due to the absence of eager
learning at home. Second, as Dewey pointed out, there may not be much potentiality for
growth in some problem which actually bothers a child. So the teacher must try to divert
him on to some other problem. Teachers come to know from experience, on the grapevine,
or from books, which topics present rich possibilities for a project. So those going round
schools find that, miraculously, children in many different schools seem to be bothered
about water, costume, or flight! So projects can become as standardized and externally
imposed as straightforward instruction.
This, on a broader view than Deweys, is not a damaging objection to the use of projects
and the problem-solving method generally; for one of the gifts of the teacher is to stimulate
interest and to get pupils to regard as problematic situations which they never previously
viewed in this light. What Dewey called interaction is not just a function of existing
internal conditions within the pupil. There is the problem, too, that if existing internal

conditions are taken as seriously as the desirability of co-operation, a highly individualized curriculum would be the result which would require something like the very favourable teacher-pupil ratio of the Dewey school and which might also mean that many pupils
end up with vast gaps in their knowledge.
The same idealistic outlook is evident in Deweys treatment of the problem of social
control, which, for many teachers nowadays, is a constant source of strain. When talking of
the problem of unruly children Dewey remarks that: There are likely to be some who, when
they come to school, are already victims of injurious conditions outside of the school.33
But this does not apply just to odd individuals; it applies also to a vast multitude of children
who come to school with an attitude towards learning which makes it very difficult for the
teacher to contrive a situation in the classroom that approximates to a game in which they
eagerly participate. Similarly the one thing which they expect of the teacher is that he or
she will be able to keep order. The attitude towards authority, which is determined by the
control system of their homes, makes it very difficult for them to take seriously a teacher
who regards himself or herself just as a friendly guide presiding to ensure continuity in
their shared experiences of problem-solving. It takes a very skilful teacher to resocialize
such children so that they are ready to learn in the way in which Dewey approved. I mention this rather mundane criticism because I do not think Dewey conceived of himself as
putting forward an ideal which could only be realized in a school like his own Laboratory
School, which catered for children whose home backgrounds were rich in experience and
favourable towards learning. His message was that the school could transforrn society; so
I think he thought that this type of learning situation could be generalized straight away. It
is at this point that my scepticism grows.
Another defect in Deweys treatment of the social medium and of his slurring over
the dualism between the childs internal conditions and what he gets from others is his
dismissal of imitation as being of much importance.34 If this is extended to include identification it can work both negatively and positively for the teacher. The negative aspect is that
so many of the models in society, with whom children identify, are anti-educational from
Deweys point of view. He always argued that education provides its own ends; it is not
merely a means to money, prestige or a good job. Yet the child is constantly presented with
models of people who have got on in the world. The questions What is the pay off? or
Where does this get you? are asked about almost anything. It is very tempting, therefore,
for the teacher to make use of extrinsic motivations such as marks, prizes, competition,
etc., perhaps refined by Skinnerian techniques, in order to get children to learn, for, because
of the ethos of individualistic societies, and the models which they throw up, these forms of
motivation are readily understood. Co-operating with others in shared experiences because
of their intrinsic value has not, unfortunately, the same straightforward appeal.
On the positive side, however, imitation and identification can work for the teacher if he
or she has mastery of and enthusiasm for what is being taught. A good example of this is
the case of Oakeshott and the sergeant gymnastics instructor mentioned at the start. Dewey,
I suspect, was hostile to imitation because it smacks of external imposition. But it is rather
cavalier for a thinker with an evolutionary orientation like Dewey to disregard one of the
main mechanisms which the human race has evolved for the transmission of culture. Bronfenbrenner, for instance, in his Two Worlds of Childhood,35 contrasts the USA and USSR

from the point of view of the degree to which systematic modelling is encouraged. His
chapter on The unmaking of the American child makes very sobering reading.
Dewey was well aware of the features of industrial society that were inimical to his
whole conception of democracy and of education. In his Individualism, Old and New36
he pointed out the irrelevance of the old individualistic values that had characterized the
pioneer; for the present problem was not that of wrestling with physical nature but that of
dealing with social conditions. Earlier individualism had shrunk to industrial initiative and
ability in making money. This was the main enemy. What was needed was a new individualism. Dewey refrained from sketching what it would be like but suggested that technology, taken in its broadest sense, offers the main clue to its nature, for it would help both to
transform society and to develop a new type of individual mind.
(iii) The technological man

What, then, is to be the verdict on Deweys ideal of the technological, problem-solving
man which is central to understanding his convictions about the methods and content of
education and his conception of democracy? Surely what was said of Bernard Shaw: He
is like the Venus de Milo. What there is of him is admirable.
There are two respects in which Deweys ideal speaks very much to our condition. First,
the plea for the use of practical intelligence, backed up by the use of science, to tackle
social and economic problems, is as pertinent today as it was at the time at which he was
writing. Second, his emphasis on shared experiences and communication and his attack
on the relics of the old individualism are apposite in a society dominated by frustration of
the desire for material gain. We could do with more fraternity, the forgotten ideal of the
French Revolution. But he was mistaken in thinking that the ideals of individualism have
shrunk just to the desire for profit. There are also autonomy, integrity, and authenticity
which are still potent individualistic ideals both in life and in education. His playing down
of such ideals is surprising; for they can scarcely be dismissed as facets of the rottenness
of perfecting an inner personality. He says of aims of education that we do not emphasize
things which do not require emphasis. He may well have thought this about such individualistic ideals, though I suspect that he did not.
(a) Neglect of the personal and of the education of the emotions What, then, are the
defects of this ideal? Mainly that, rather ironically, in putting forward an ideal which is
meant to resolve current dualisms, he develops a very onesided view of man that completely
ignores certain features of the human condition. First, Dewey ignores the purely personal
life of human beings. By that I dont mean just his failure to emphasize the importance of
respect for persons in his account of democracy, nor his attack on the rottenness of individual attempts at self-improvement; I mean also his neglect of interpersonal relationships
and the education of the emotions. It is significant that he makes practically no mention of
the role of literature in education. Literature is singularly unamenable to the problem-solving method of learning, and often concerns itself with the predicaments of man rather than
with his problems. I once attended a poetry lesson in an American school. The teacher read
part of Grays Elegy beautifully and then opened up a somewhat desultory discussion on
what a curfew was, etc. I asked her why she did not read it again as she had such a lovely
voice and had the pupils spellbound. Oh, she said, we are only allowed to read it once.

It is meant to provide material for discussion and problem-solving to help the children to
become democratic citizens. She was surely echoing John Deweys ideology.
(b) Predicaments as well as problems This brings me to the second comment on Deweys idealthe emphasis on problem-solving. He shared the view of most Americans of
that period that life presented mainly problems that could be solved, given the time and the
technology. This optimism is of course waning somewhat, in the face of the intractability
of problems connected with race, unemployment, and poverty. But even if it were not, the
view of life presented is onesided, not to say exhausting. Dewey, of course, appreciated the
importance of habit in life, but accorded no value to anything that was a matter of routine.
Yet there is nothing particularly wicked about the conservative pleasures derived from repetition and familiarity. There are also the more distanced, aesthetic enjoyments that have
little to do with problem-solving. There are many aspects of life, too, that present not problems that can be solved but predicaments that have to be lived with. If a man in his prime
is afflicted by a coronary or loses his wife, he does not just have a problem.
(c) Disregard of the irrational The third criticism is like the second in that it is directed
against Deweys confident, reformist optimism. He completely ignores the fundamental
irrationality of man. He never mentions Freud, who was a contemporary of his, and seems
sublimely unaware of the diagnosis of the human condition that derived from his insights.
The view that civilization is a brittle crust containing with difficulty irrational yearnings,
made no impact on Dewey in spite of his active interest in the rise of Nazism as a threat to
democracy37and what a limited perspective on Nazism!
(d) Defects of the pragmatic stance There are finally the defects of Deweys pragmatic
stance. Others have commented in detail on the defects of the pragmatic theory of truth.38
I shall confine myself to more general issues, which are crucial to his educational ideal.
Basically the pragmatist lacks reverence, is guilty of what Russell calls cosmic impiety.39
He sees nature just as something that can be used for human purposes. He lacks a sense of
awe and of wonder. This is manifest, too, in Deweys insistence that history and geography
must be taught subject to the condition that they throw light on contemporary problems
and concerns. There is no reason why these should not be used as a starting point if they
are motivationally potent, but to view such studies only under this aspect is both to distort
them and to encourage a kind of present-centred
It is the same with science. Dewey actually grossly exaggerated the connection between
scientific theories and everyday practical problems. But to represent scientific theories,
which are some of the greatest products of the human imagination, just as aids to action, is
to ignore a whole dimension of human life. His psychology is made to fit this emphasis; for
the child is credited with an impulse to investigate and experiment but not with a more
generalized drive to know with which modern psychologists credit even monkeys. So
Dewey put into the child at the beginning in the form of impulses what he took out at the
end in the form of the co-operative, communicative, technological man. The dimensions of
speculative curiosity, of wonder and awe, are missing.
I am not of course suggesting that technology is unimportant in comparison with disinterested speculation. Still less am I suggesting that relevance is an unimportant criterion
of learning and of the curriculum. What I am suggesting is that Deweys ideal is as myopic
as his conception of relevance. Contributing to practical purposes is only one criterion of
relevance. The others are not limited to that which arouses plain curiosity; there are also

countless studies in literature, religion, history, psychology, and the social sciences which
are of great emotional significance to human beings without being obviously connected
with practical purposes. When Whitehead said that education is the acquisition of the art
of the utilization of knowledge40 Im sure that he did not think of utilization purely in
terms of relevance to practical purposes. What he meant was that the content of education
should have application to peoples lives. It should not consist in inert ideas propounded
by teachers on the assumption that their pupils are going to be devotees of their subject like
themselvesfor one of the crucial questions for any teacher is what there is in his subject
for the majority who are unlikely to become specialists in it.
(e) Conclusion To sum up: Deweys revolt against the formalism and irrelevance of
much that went on in schools is still pertinent. So is his plea for more shared experiences
and more development of practical intelligence. But his ideal of the technological man
is too limited and culture bound. It ignores whole dimensions of the human condition
especially the predicaments of man, his irrationality, and his emotional sensitivities and
susceptibilities. The cult of co-operative action is a welcome antidote to the lonely quest
for salvation or for private profit. But human beings inhabit a personal as well as a public
world; they are circumscribed by a Nature that has to be accepted as well as transformed,
that should be an object of enjoyment, of wonder and of awe as well as material to be
mastered for human purposes. A balance has to be struck between personal preoccupations
and public policies, between servile humility and masterful
These are dualisms that
Dewey did not resolve.
Notes
My thanks are due to my colleagues, Pat White and Robert Dearden, for their helpful comments on a first revision of this paper.
1 Oakeshott, M., Learning and teaching, in R.S.Peters (ed.), The Concept of Education, Routledge
& Kegan Paul, London, 1967.
2 Dewey, J., The Child and the Curriculum, University of Chicago Press, 1902.
3 Dewey, J., The School and Society, University of Chicago Press, 1900; rev. eds, 1915, 1943.
4 Dewey, J., Interest and Effort in Education, Houghton Mifflin, Boston, 1913.
5 Dewey, J., Experience and Education, Macmillan, New York, 1938.
6 Dewey, J., Democracy and Education, Macmillan, New York, 1916.
7 Ibid., ch. VII.
8 Ibid., p. 143.
9 Bradley, F.H., Ethical Studies, Oxford University Press, 1876, ch. V.
10 Op. cit.
11 Op. cit., pp. 8990.
12 Dewey, J., How We Think, Heath, Boston, 1910.
13 Dearden, R.F., The Philosophy of Primary Education, Routledge & Kegan Paul, London, 1968.
14 Murdoch, I., The Sovereignty of Good, Routledge & Kegan Paul, London, 1970.
15 Dewey, J., Human Nature and Conduct, Henry Holt, New York, 1922.
16 Dewey, J., Democracy and Education, Macmillan, New York, 1916, p. 127.
17 Ibid., pp. 412.
18 Op. cit.
19 Ibid., p. 21.
20 Mayhew, K.C. and Edwards, A.C., The Dewey School, Atherton Press, New York, 1966, p. 22.

21 Ibid., pp. 46,41.
22 Dewey, J., Experience and Education, Macmillan, New York, 1938, p. 59.
23 Dewey, J., The School and Society, University of Chicago Press, 7th impression, 1963, p. 29.
24 Ibid., p. 22.
26 Ibid., p. 251.
27 Ibid., p. 263.
28 Dewey, J., The School and Society, University of Chicago Press, 7th impression, 1963, p. 91.
29 Dykhuizen, G., The Life and Mind of John Dewey, Southern Illinois University Press, 1973, pp.
1413.
31 Hook, S., Education and the Taming of Power, Open Court, New York, 1973.
32 Dewey, J., Reconstruction in Philosophy, Henry Holt, New York, 1920, p. 211.
33 Dewey, J., Experience and Education, Macmillan, New York, 1938, p. 56.
34 Dewey, J., Democracy and Education, Macmillan, New York, 1916, pp. 403.
35 Bronfenbrenner, U., Two Worlds of Childhood, Allen & Unwin, London, 1971.
36 Dewey, J., Individualism, Old and New, Minton, Baker, New York, 1936.
37 See Dykhuizen, op. cit., pp. 2401.
38 See, for instance, I.Scheffler, Four Pragmatists, Routledge & Kegan Paul, London, 1974.
39 Russell, B., A History of Western Philosophy, Allen & Unwin, London, 1946, p. 856.
40 Whitehead, A.N., The Aims of Education, Macmillan, London, 1921; reprinted Mentor Books,
USA, 1949, p. 16.
Index
Note: the index does not include authors of works referred to only in bibliographical
references
accomplishment, 29, 31
action and behaviour, 467, 57, 6174;
education and, 105, 120;
interest and, 38, 4450, 71;
knowledge and, 89, 1214, 120;
language and, 20, 234, 313;
pointing and, 27
Allport, G.W., 57
Anderson, J., 93
Arblaster, A., 98
Aristotle, 11, 49, 51, 53
Armstrong, D., 93
Arnold, F., 35, 42
assessment (in education), 97
association, 59
attention, 234, 278;
interest and, 3541, 434, 4650, 53
Austin, J., 23, 76, 98
authority (testimony), 1617
Ayer, A.J., 1, 71
Baldwin, J.M., 44
Bantock, G.H., 868
behaviour, see action
belief, see concepts;
knowledge
Berkeley, G., 1
Berlyne, D.E., 35, 44
Bloch, O., 31
Bloom, L., 30, 323
Bradley, F.H., 104
Bronfenbrenner, U., 117
Buddhism, 71
Cartesianism, 18;
see also Descartes
causation, 15, 65, 67
change, 1314
child:
acquisition of language by, 1833;
Deweys concept of, 115, 120;
interaction with mother, 225, 2733;
as pupil, see learning;
school;
teaching
choice, 63, 66, 72
Chomsky, N., 1822
class, social, 878, 99
classification, disposition for, 1112
communication, interpersonal, 11, 17, 1920,
91, 103, 1056, 111;
childs mastery of, 30, 323
conatus, 73
concepts, formation of, 912, 22, 31, 71;
see also knowledge;
mind;
thought
conditioning, 61, 65
consciousness, 35, 424, 59, 69;
see also mind;
self;
thought
Copernicus, N., 113
curiosity, 9, 47, 1201
curriculum, 10912, 11921
Dearden, R., 105, 122
DeLaguna, G., 29, 32
democracy, 72, 96;
definition, 7681, 114;
Deweys view of, 817, 93, 989, 103, 106,
109, 112, 11415, 11718
Descartes, R., 59, 61, 105;
see also Cartesianism
desire, 423, 46
Dewey, J.:
80 Index
on education, 10221;
on experience, 1920;
on interest, 3554;
on knowledge, 28, 1214, 17, 5860;
on language, 1820, 30, 33;
on society, 5674, 8196, 989, 10221
dualism, Deweys attempts to resolve, 1047,
11118, 121
Durkheim, E., 63
education, 10221;
content of, 10912, 11921;
control of, 95, 100;
democracy and, 803, 87, 93, 98, 106, 114,
11718;
freedom and, 58, 62;
functions of, 87, 956, 99;
interest and, 43, 534, 108;
see also learning;
school;
teaching
Education for Democracy (Rubinstein and
Stoneman), 76, 803, 95, 978
Edwards, A.C., 108
elitism, 80, 83
emotion, 424, 119, 121
environment:
control of, 106, 111;
interaction of person with, 34, 1112,
1416, 934;
see also nature;
object;
self
equality, 845;
of opportunity, 88, 99;
of outcome, 878
exchange routine, 27, 2931
experience, 67, 579, 61, 71, 105, 108;
language and, 1823, 323;
see also experiment;
perception
experiment, 8, 1214
fallibilism, 38
feeling, interest and, 356, 424
Ferrier, L.J., 22
freedom, 578, 613, 66, 73, 84, 105
Freud, S., 120
Gallie, W.B., 2
Goffman, E., 70
grammar, 201, 313
Greenfield, P., 30
Grice, H.P., 26
growth, 1046, 113, 115
habit, 57, 613, 667, 713, 119
Halliday, M., 23
Heading, B., 74
Hegel, G.W.F., 578, 102, 1067, 109
Herbart, J.F., 35, 44
heredity, 878
Hobbes, T., 66, 71, 73
Homer, 10
Hook, S., 93, 100, 113;
quoted, 8990
Hume, D., 12, 5, 5760, 71, 91
Husserl, E.G.A., 70
Huxley, R., 33
identity, individual, see self
imagination, 59
imitation, 29, 33, 107, 117
impulse, 667, 73, 115, 120
individual, see self and under society
individualism, 11718
inference, 67, 16
inquiry, 89, 1214
instrumentalism, 814, 59, 71
intellectualism, 34, 9, 1112
intention, 29, 31, 934
interactionism, symbolic, 63, 69
interest and interests:
action and, 38, 4450, 71;
attention and, 3541, 434, 4650, 53;
education and, 43, 534, 108;
feeling and, 356, 424;
motives and, 446, 67;
needs and, 11, 43;
objects of, 356, 42, 445, 503;
in politics, 789;
in society, 81, 837, 90, 92, 103
James, W., 33, 35, 102;
on self, 57, 602, 66, 69, 71, 74;
on truth, 12
Kador, J., 789
Kant, I., 11, 74, 104, 114
Index 81
Kepler, J., 113
Kharume, A., 79
Kilpatrick, W.H., 108
knowledge:
action and, 89, 1214, 120;
belief and certainty in, 38;
education and, 956, 105, 121;
language and, 23;
theories of, 117
labelling, 235, 2731, 33
language, 10, 18, 91;
acquisition of, 1833;
functions of, 23, 303
learning, 48, 93, 97, 102, 108, 11113, 11517,
11921;
of language, 1833;
see also school;
teaching
Leibniz, G.W., 70
Lenin, V.I., 97
literature, 119
Locke, J., 1
McDougall, W., 37
Mach, E., 2
Mclntyre, D., 97
Mackie, J., 93
McNeill, D., 323
Mandeville, B. de, 92
Mauger, P., 99
Mayhew, K.C., 108
Mead, G.H., 601, 71, 1067
meaning, 202, 63, 65, 6970, 89
memory, 6, 60, 71
Mill, J.S., 1, 35, 95, 100
Miller, G., 20
mind, 34, 1112, 14, 16, 35, 5860;
see also concepts;
consciousness;
self;
thought
mood, 43
mother, interaction with child, 225, 2733
motive, 35, 446, 49, 647, 74;
for learning, 117
Nagel, E., 93
nature, 58, 1045, 107, 1201;
human, see self;
see also environment

Nazism, 120
negation, 32
Oakeshott, M., 102, 117
object (of interest), 356, 42, 445, 503;
see also environment
Passmore, J., 93
Peirce, C., 12, 6, 102
perception, 6, 11, 1416, 59
personality, see self
Piaget, J., 114
pictures, 278
Plato, 87
Platonism, 9, 11
pointing (by infants), 2330, 33
politics, 7782, 86, 989;
see also democracy;
society
Popper, K., 2
pragmatism, 12, 567, 5960, 1023, 120
predication, 23
Price, H.H., 1, 15
Quine, W.V.O., 10
rationality, human nature and, 612, 67, 714,
120
reality, 58, 634, 701
relevance (in education), 121
response, conditioned, 61, 65
role, social, 6970;
see also society (individual in)
Rousseau, J.J., 72, 79, 107
Russell, B., 12, 5, 18, 120
Ryle, G., 49, 53
Schiller, F.C.S., 2
school:
home and, 11112, 11516;
in society, 10917;
see also learning;
teaching
science, nature of, 810, 120
selection (in education), 83
self (individual identity, personality), 35, 45,
523, 89, 1036, 11821;
individual autonomy, 105;
theories of, 5664, 6674;
82 Index
see also consciousness;
mind;
and under society
Shaw, G.B., 118
Skinner, B.F., 61, 117
Skinner, Q., 74
Smith, A., 92
Smith, J.H., 30
society:
individual in, 578, 604, 6770, 89, 912,
1037, 118, 121;
institutions of, 72, 74
(see also democracy);
school in, 10917;
see also under Dewey
Spinoza, B. de, 73
Stoics, 11
Stout, G.F., 35, 37
Stumpf, K., 35, 44
syntax, 1823
task structure, 31
teaching and teacher, 38, 40, 48, 93, 957,
1069, 11417;
see also learning;

school
technology, 11821
testimony (authority), 1617
thought, 1920, 61, 668, 71;
see also concepts;
consciousness;
mind
Titchener, E.B., 35
Tocqueville, A. de, 80
translation, 10
vocalization, see language (acquisition of)
Warnock, G.J., 15
Weber, M., 63, 65
White, P., 122
Whitebread, N., 823
Whitehead, A.N., 18, 121
Whitman, W., 96
Wilson, P.S., 35
Wittgenstein, L., 89

Phileducation John Dewey

Transféré par

Informations du document

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Phileducation John Dewey

Transféré par

Droits d'auteur :

Formats disponibles

INTERNATIONAL LIBRARY OF THE

JOHN DEWEY RECONSIDERED

JOHN DEWEY RECONSIDERED

LONDON AND NEW YORK

First published in 1977

ISBN 10:0-415-55946-4 (Set)

John Dewey reconsidered

Routledge & Kegan Paul

First published in 1977

ISBN 0 7100 8623 7 (Print Edition)

General editors note

1 Inquiry, thought and action: John Deweys theory of knowledge

2 Language and experience

3 Deweys theory of interest

4 The self in action

5 Democracy and education

6 John Deweys philosophy of education

General editors note

There is a growing interest in philosophy of education amongst students of philosophy as

viii General editors note

2 John Dewey reconsidered

2 The four themes of anti-intellectualism

Inquiry, thought and action: John Deweys theory of knowledge 3

4 John Dewey reconsidered

Inquiry, thought and action: John Deweys theory of knowledge 5

6 John Dewey reconsidered

Inquiry, thought and action: John Deweys theory of knowledge 7

8 John Dewey reconsidered

Inquiry, thought and action: John Deweys theory of knowledge 9

10 John Dewey reconsidered

6 The social nature of knowledge

Inquiry, thought and action: John Deweys theory of knowledge 11

Language and experience 13

14 John Dewey reconsidered

Language and experience 15

16 John Dewey reconsidered

Language and experience 17

18 John Dewey reconsidered

Language and experience 19

20 John Dewey reconsidered

Language and experience 21

22 John Dewey reconsidered

24 John Dewey reconsidered

Deweys theory of interest 25

26 John Dewey reconsidered

Deweys theory of interest 27

28 John Dewey reconsidered

Deweys theory of interest 29

30 John Dewey reconsidered

Deweys theory of interest 31

32 John Dewey reconsidered

Deweys theory of interest 33

34 John Dewey reconsidered

Deweys theory of interest 35

The self in action 37

38 John Dewey reconsidered

The self in action 39

40 John Dewey reconsidered

The self in action 41

42 John Dewey reconsidered

The self in action 43

44 John Dewey reconsidered

The self in action 45

46 John Dewey reconsidered